WO2017104617A1 - ガス検知用画像処理装置、ガス検知用画像処理方法、ガス検知用画像処理プログラム、ガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、及び、ガス検知システム - Google Patents
ガス検知用画像処理装置、ガス検知用画像処理方法、ガス検知用画像処理プログラム、ガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、及び、ガス検知システム Download PDFInfo
- Publication number
- WO2017104617A1 WO2017104617A1 PCT/JP2016/086929 JP2016086929W WO2017104617A1 WO 2017104617 A1 WO2017104617 A1 WO 2017104617A1 JP 2016086929 W JP2016086929 W JP 2016086929W WO 2017104617 A1 WO2017104617 A1 WO 2017104617A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- gas
- value
- pixel
- identification
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/38—Investigating fluid-tightness of structures by using light
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/002—Investigating fluid-tightness of structures by using thermal means
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
Definitions
- the present invention relates to a technique for detecting gas using an infrared camera.
- Patent Document 1 includes an infrared camera that captures an inspection target region, and an image processing unit that processes an infrared image captured by the infrared camera, and the image processing unit includes a plurality of time-series arranged.
- a gas leak detection apparatus having a fluctuation extraction unit that extracts dynamic fluctuation due to gas leak from an infrared image is disclosed.
- Patent Document 2 includes a first infrared camera that measures the intensity of infrared light in the first wavelength range that is absorbed by the detection target gas, and a wavelength range different from the first wavelength range and a first wavelength range.
- a detection target gas based on the second infrared camera that measures the intensity of infrared light in the second wavelength range, the result measured by the first infrared camera, and the result measured by the second infrared camera And a determination unit for determining the presence or absence of the gas.
- a gas detection device that generates a monitoring image indicating a leaked gas as a gas image using an infrared image of a monitoring target of gas leakage, and displays it on a display unit.
- a portion where a slight temperature change due to the leaked gas occurs can be visualized and displayed as a gas image.
- the image of the moving object the movement of a person, the movement of grass, trees, etc.
- the gas image and the moving object It is necessary to make sure that the image is correct.
- the present invention relates to a gas detection image processing apparatus, a gas detection image processing method, and a gas detection image processing capable of discriminating between a gas image and a non-gas image in a monitoring image generated using an infrared image to be monitored. It is an object to provide a computer-readable recording medium in which a program, a gas detection image processing program are recorded, and a gas detection system.
- the gas detection image processing apparatus includes a calculation unit and an identification unit.
- the calculation unit uses the identification value corresponding to each of a plurality of pixels constituting an infrared image to be monitored for an identification value for identifying a pixel constituting a gas image and a pixel constituting a non-gas image. calculate.
- the identification unit identifies a pixel constituting the gas image and a pixel constituting the non-gas image in the monitoring image generated using the infrared image based on the identification value.
- FIG. 1 is a block diagram of a gas detection system including an image processing apparatus for gas detection according to the present embodiment. It is a block diagram which shows the hardware constitutions of the image processing apparatus for gas detection shown to FIG. 1A. It is explanatory drawing explaining time series pixel data. It is a flowchart explaining the 1st aspect of this embodiment. It is an image figure which shows the infrared image which image
- gas temperature is lower than background temperature, it is explanatory drawing explaining the relationship of gas temperature, background temperature with gas, and background temperature without gas.
- gas temperature is higher than background temperature, it is explanatory drawing explaining the relationship of gas temperature, background temperature with gas, and background temperature without gas.
- FIG. 6 is a first explanatory diagram illustrating correction of a temperature difference between a gas temperature and a gas-free background temperature in a correlation value.
- FIG. 10 is a second explanatory diagram illustrating correction of a temperature difference between a gas temperature and a gas-free background temperature in the correlation value. It is an image figure which shows the various images of the test place image
- FIG. 1A is a block diagram of a gas detection system 1 including a gas detection image processing apparatus 3 according to the present embodiment.
- the gas detection system 1 includes an infrared camera 2 and a gas detection image processing device 3.
- the infrared camera 2 captures a moving image of an infrared image of an object to be monitored for gas leakage (for example, a place where gas transport pipes are connected) and a background, and generates moving image data D1 indicating the moving image.
- the moving image is an example of a plurality of infrared images arranged in time series. Not only a moving image but also an infrared camera 2 may be used to capture a gas leak monitoring target and a background infrared image at a plurality of times.
- the infrared camera 2 includes an optical system 4, a filter 5, a two-dimensional image sensor 6, and a signal processing unit 7.
- the optical system 4 forms an infrared image of the subject (monitoring target and background) on the two-dimensional image sensor 6.
- the filter 5 is disposed between the optical system 4 and the two-dimensional image sensor 6, and allows only infrared light having a specific wavelength to pass through the light that has passed through the optical system 4.
- the wavelength band that passes through the filter 5 depends on the type of gas to be detected. For example, in the case of methane, a filter 5 that passes a wavelength band of 3.2 to 3.4 ⁇ m is used.
- the two-dimensional image sensor 6 is a cooled indium antimony (InSb) image sensor, for example, and receives the infrared rays that have passed through the filter 5.
- the signal processing unit 7 converts the analog signal output from the two-dimensional image sensor 6 into a digital signal and performs known image processing. This digital signal becomes the moving image data D1.
- the moving image indicated by the moving image data D1 has a structure in which a plurality of frames are arranged in time series. Data obtained by arranging pixel data of pixels at the same spatial position in time series in a plurality of frames is set as time series pixel data.
- the time series pixel data will be specifically described.
- FIG. 2 is an explanatory diagram for explaining time-series pixel data.
- K be the number of frames of a moving image of an infrared image. Assume that one frame includes M pixels, that is, a first pixel, a second pixel,..., An M ⁇ 1th pixel, and an Mth pixel.
- the pixel data indicates the luminance or temperature of the pixel.
- pixels in the same spatial position mean pixels in the same order.
- the pixel data of the first pixel included in the first frame the pixel data of the first pixel included in the second frame,..., The (K ⁇ 1) th frame.
- the data obtained by arranging the pixel data of the first pixel included in the pixel data and the pixel data of the first pixel included in the Kth frame in time series becomes the time series pixel data of the first pixel.
- the number of time-series pixel data is the same as the number of pixels constituting one frame, and the moving image data D1 is composed of the plurality (M) of time-series pixel data.
- the gas detection image processing device 3 is a personal computer, a smartphone, a tablet terminal, or the like, and includes, as functional blocks, an image generation unit 8, an arithmetic processing unit 9, a display control unit 12, a display unit 13, An input unit 14 and an image data input unit 15 are provided.
- the image data input unit 15 is a communication interface that communicates with a communication unit (not shown) of the infrared camera 2.
- the video data D1 sent from the communication unit of the infrared camera 2 is input to the image data input unit 15.
- the moving image data D1 is an example of image data.
- the image data is data indicating a plurality of infrared images obtained by photographing a gas leak monitoring target at a plurality of times.
- the image data input unit 15 sends the moving image data D1 to the image generation unit 8, the arithmetic processing unit 9, and the display control unit 12.
- the image generation unit 8, the arithmetic processing unit 9, and the display control unit 12 are realized by a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), and the like. .
- a CPU Central Processing Unit
- RAM Random Access Memory
- ROM Read Only Memory
- HDD Hard Disk Drive
- the image generation unit 8 performs predetermined image processing on the moving image data D1 to generate a predetermined image (for example, a monitoring image).
- the calculation processing unit 9 performs various calculations necessary for gas detection image processing.
- the arithmetic processing unit 9 includes a calculation unit 10 and an identification unit 11. These will be described later.
- the display control unit 12 causes the display unit 13 to display a predetermined image generated by the image generation unit 8.
- the display unit 13 is realized by a liquid crystal display, for example.
- the input unit 14 is realized by a keyboard or a touch panel, and performs various inputs related to gas detection.
- FIG. 1B is a block diagram showing a hardware configuration of the gas detection image processing apparatus 3 shown in FIG. 1A.
- the gas detection image processing apparatus 3 includes a CPU 3a, a RAM 3b, a ROM 3c, an HDD 3d, a liquid crystal display 3e, a communication interface 3f, a keyboard 3g, and a bus 3h for connecting them.
- the liquid crystal display 3 e is hardware that implements the display unit 13. Instead of the liquid crystal display 3e, an organic EL display (Organic Light Emitting Diode display), a plasma display, or the like may be used.
- the communication interface 3 f is hardware that implements the image data input unit 15.
- the keyboard 3 g is hardware for realizing the input unit 14.
- the HDD 3d stores programs for realizing the functional blocks of the image generation unit 8, the arithmetic processing unit 9, and the display control unit 12, respectively.
- the program that realizes the image generation unit 8 is a processing program that acquires moving image data D1 (image data) and performs the predetermined processing (for example, monitoring image generation processing) on the moving image data D1.
- the program for realizing the arithmetic processing unit 9 is an arithmetic program for performing various calculations necessary for the image processing for gas detection.
- the program that realizes the display control unit 12 is a display control program that causes the display unit 13 to display an image (for example, a predetermined image generated by the image generation unit 8). These programs may be stored in the ROM 3c instead of the HDD 3d.
- the CPU 3a reads out the processing program, the arithmetic program, and the display control program from the HDD 3d, expands them in the RAM 3b, and executes the expanded programs to realize these functional blocks.
- the processing program, the arithmetic program, and the display control program are stored in advance in the HDD 3d, but are not limited thereto.
- a recording medium for example, an external recording medium such as a magnetic disk or an optical disk
- these programs may be stored in the HDD 3d.
- These programs may be stored in a server connected to the image processing apparatus 3 for gas detection via a network, and these programs may be sent to the HDD 3d via the network and stored in the HDD 3d.
- the gas detection image processing apparatus 3 has a first mode to a fourth mode as described below.
- Each of these aspects is constituted by a plurality of elements.
- the HDD 3d stores a program for realizing these elements.
- the first aspect of the gas detection image processing apparatus 3 includes a calculation unit and an identification unit as elements.
- the HDD 3d stores programs for realizing each of the calculation unit and the identification unit. These programs are expressed as a calculation program and an identification program.
- the calculation unit and calculation program will be described as an example.
- the calculation unit calculates an identification value corresponding to each of the plurality of pixels constituting the monitoring target infrared image, with respect to the identification value for identifying the pixel constituting the gas image and the pixel constituting the non-gas image.
- the calculation program calculates an identification value corresponding to each of a plurality of pixels constituting the monitoring target infrared image with respect to an identification value for identifying a pixel constituting the gas image and a pixel constituting the non-gas image. It is a program.
- FIG. 3 is a flowchart illustrating the first aspect of the present embodiment.
- the monitoring image generation process step S100
- the gas concentration / thickness product calculation process step S101
- the pixels constituting the gas image included in the monitoring image and the gas image included in the monitoring image are displayed. It is comprised by the identification process (step S102) with the pixel which comprises the image of moving bodies other than.
- an image of a moving object other than a gas image may be simply referred to as an “image of a moving object”.
- the monitoring image is an image generated using an infrared image of the monitoring target, and includes a gas image in which the leaked gas is visualized when gas leaks from the monitoring target.
- the gas concentration thickness product (hereinafter sometimes simply referred to as “concentration thickness product”) will be described.
- concentration thickness product When a gas leak is detected, the risk level of the gas (for example, the possibility of explosion) needs to be determined.
- the risk of gas can be determined by the gas concentration at the location where the gas is drifting.
- remote gas detection using an infrared camera cannot directly measure the gas concentration at the location where the gas is drifting, but measures the gas concentration thickness product.
- the gas concentration / thickness product means a value obtained by integrating the gas concentration along the depth direction of the portion where the gas drifts.
- Monitoring image generation processing There are various methods for generating a monitoring image. Here, an example of a method for generating a monitoring image will be described.
- the monitoring image is generated using an infrared image of the monitoring target and the background.
- the present inventor found that a gas leak and a background temperature change occur in parallel, and if the background temperature change is larger than the temperature change due to the leaked gas, the background temperature If changes are not taken into account, it has been found that the state of gas leakage cannot be displayed as an image. This will be described in detail.
- FIG. 4 is an image diagram showing, in time series, an infrared image obtained by photographing an outdoor test place in a state where gas leakage and background temperature change occur in parallel. These are infrared images obtained by shooting a moving image with an infrared camera. There is a point SP1 at the test place where gas can be ejected. For comparison with the point SP1, a point SP2 where no gas is ejected is shown.
- the image I1 is an infrared image of the test place taken at time T1 immediately before the sunlight is blocked by the clouds.
- the image I2 is an infrared image of the test place taken at time T2 after 5 seconds from time T1. At time T2, sunlight is blocked by clouds, so the background temperature is lower than at time T1.
- the image I3 is an infrared image of the test place taken at time T3 10 seconds after time T1. From time T2 to time T3, the state in which the sunlight is blocked by the cloud is continued, so that the temperature of the background is lower at time T3 than at time T2.
- the image I4 is an infrared image of the test place taken at time T4 15 seconds after time T1. From time T3 to time T4, the state in which sunlight is blocked by the cloud is continued, so that the background temperature is lower at time T4 than at time T3.
- the background temperature has dropped by about 4 ° C in 15 seconds from time T1 to time T4. For this reason, the image I4 is generally darker than the image I1, and it can be seen that the background temperature is lowered.
- FIG. 5A is a graph showing the temperature change at the point SP1 at the test location
- FIG. 5B is a graph showing the temperature change at the point SP2 at the test location.
- the vertical axis of these graphs indicates temperature.
- the horizontal axis of these graphs indicates the frame order. For example, 45 means the 45th frame.
- the frame rate is 30 fps. Therefore, the time from the first frame to the 450th frame is 15 seconds.
- the graph showing the temperature change at the point SP1 is different from the graph showing the temperature change at the point SP2. Since no gas is ejected at the point SP2, the temperature change at the point SP2 indicates the background temperature change. On the other hand, since gas is ejected at the point SP1, gas is drifting at the point SP1. For this reason, the temperature change at the point SP1 indicates the temperature change obtained by adding the background temperature change and the temperature change caused by the leaked gas.
- the moving image data D1 (FIG. 1A) has a lower frequency than the first frequency component data indicating the temperature change due to the leaked gas and a background temperature change. This is because the second frequency component data is included. The image indicated by the first frequency component data becomes invisible due to the image indicated by the second frequency component data (background light change).
- a graph indicating the temperature change at the point SP2 corresponds to the second frequency component data.
- the image generation unit 8 (FIG. 1A) generates a plurality of time-series pixel data having different pixel positions (that is, a plurality of time-series pixel data constituting the moving image data D1) from the moving image data D1.
- the second frequency component data is removed from each of the time-series pixel data.
- the plurality of time-series pixel data having different pixel positions refers to the time-series pixel data of the first pixel, the time-series pixel data of the second pixel,..., M ⁇ 1th, with reference to FIG. Means time-series pixel data of the second pixel and time-series pixel data of the Mth pixel.
- the frequency component data having a frequency higher than the frequency of the first frequency component data and indicating high frequency noise is set as the third frequency component data.
- the image generation unit 8 performs a process of removing the third frequency component data in addition to the process of removing the second frequency component data for each of the plurality of time-series pixel data constituting the moving image data D1.
- the image generation unit 8 does not perform the process of excluding the second frequency component data and the third frequency component data in units of frames, but the second frequency component data and the units of time series pixel data. Processing for removing the third frequency component data is performed.
- FIG. 6 is a flowchart illustrating the monitoring image generation process shown in step S100 of FIG.
- the time series pixel data before the predetermined processing is performed on the pixel data of the time series pixel data is set as the first time series pixel data
- the time series pixel data after the predetermined processing is set as the second time series. Let it be pixel data.
- the predetermined processing is processing for generating a monitoring image, an image in which the density / thickness product is visualized, and the like.
- Time series pixel data D2 shown in FIG. 7 described later is first time series pixel data
- third difference data D9 shown in FIG. 10 is second time series pixel data.
- the time series pixel data shown in FIG. 2 is the first time series pixel data before the pixel data is subjected to the predetermined processing, and the second time series when the pixel data is after the predetermined processing. It becomes pixel data.
- the image generation unit 8 calculates a simple moving average for the first time-series pixel data by using a first predetermined number of frames smaller than the K frames shown in FIG.
- Data extracted from the series pixel data is set as second frequency component data, and M second frequency component data corresponding to each of the M time series pixel data shown in FIG. 2 is extracted (step S1). .
- the first predetermined number of frames is, for example, 21 frames.
- the breakdown is a target frame, 10 consecutive frames before this, and 10 consecutive frames after this.
- the first predetermined number is not limited to 21 but may be more than 21 or less than 21 as long as the second frequency component data can be extracted from the time series pixel data.
- the image generation unit 8 calculates a simple moving average for the first time-series pixel data with a third predetermined number (for example, 3) frames smaller than the first predetermined number (for example, 21) as a unit.
- the data extracted from the first time-series pixel data is set as third frequency component data, and M third frequency component data corresponding to each of the M time-series pixel data shown in FIG. Is extracted (step S4).
- FIG. 7 shows the time-series pixel data D2 of the pixel corresponding to the point SP1 (FIG. 4), the second frequency component data D3 extracted from the time-series pixel data D2, and the third extracted from the time-series pixel data D2.
- the vertical axis and horizontal axis of the graph are the same as the vertical axis and horizontal axis of the graph of FIG. 5A.
- the temperature indicated by the time-series pixel data D2 changes relatively abruptly (change period is relatively short), and the temperature indicated by the second frequency component data D3 changes relatively slowly (change). Is relatively long).
- the third frequency component data D6 appears to overlap with the time-series pixel data D2.
- the third predetermined number of frames is, for example, 3 frames.
- the breakdown is the target frame, the immediately preceding frame, and the immediately following frame.
- the third predetermined number is not limited to 3 and may be more than 3 as long as the third frequency component can be extracted from the time-series pixel data.
- the image generation unit 8 uses the data obtained by calculating the difference between the first time-series pixel data and the second frequency component data extracted from the first time-series pixel data as the first difference data. , M pieces of first difference data corresponding to each of the M pieces of time-series pixel data are calculated (step S2).
- the image generation unit 8 uses the data obtained by calculating the difference between the first time-series pixel data and the third frequency component data extracted from the first time-series pixel data as the second difference data. , M pieces of second difference data corresponding to each of the M pieces of time-series pixel data are calculated (step S5).
- FIG. 8A is a graph showing the first difference data D4, and FIG. 8B is a graph showing the second difference data D7.
- the vertical and horizontal axes of these graphs are the same as the vertical and horizontal axes of the graph of FIG. 5A.
- the first difference data D4 is data obtained by calculating a difference between the time-series pixel data D2 and the second frequency component data D3 shown in FIG.
- the repetition of the minute amplitude indicated by the first difference data D4 is mainly the sensor of the two-dimensional image sensor 6. Noise is shown. After starting the gas ejection at the point SP1 (the 90th and subsequent frames), variations in the amplitude and waveform of the first difference data D4 are large.
- the second difference data D7 is data obtained by calculating a difference between the time-series pixel data D2 and the third frequency component data D6 shown in FIG.
- the first difference data D4 includes first frequency component data (data indicating a temperature change due to leaked gas) and third frequency component data D6 (data indicating high-frequency noise).
- the second difference data D7 does not include the first frequency component data but includes the third frequency component data D6.
- the amplitude and waveform variations of the first difference data D4 vary after starting the gas ejection at the point SP1 (the 90th and subsequent frames). It is getting bigger.
- the second difference data D7 does not include the first frequency component data, and thus does not occur.
- the second difference data D7 repeats a minute amplitude. This is high frequency noise.
- the first difference data D4 and the second difference data D7 are correlated, but not completely correlated. That is, in a certain frame, the value of the first difference data D4 may be plus and the value of the second difference data D7 may be minus or vice versa. For this reason, even if the difference between the first difference data D4 and the second difference data D7 is calculated, the third high-frequency component data D6 cannot be removed. In order to remove the third high-frequency component data D6, it is necessary to convert the first difference data D4 and the second difference data D7 into values such as absolute values that can be subtracted.
- the image generation unit 8 uses the first variation to obtain data obtained by calculating a moving standard deviation with a second predetermined number of frames smaller than K frames as a unit.
- M pieces of first variation data corresponding to each of the M pieces of time-series pixel data are calculated (step S3).
- the movement variance may be calculated instead of the movement standard deviation.
- the image generation unit 8 obtains data obtained by calculating a moving standard deviation in units of a fourth predetermined number (for example, 21) frames smaller than K frames with respect to the second difference data. Then, M second variation data corresponding to each of the M time-series pixel data are calculated as the second variation data (step S6).
- moving standard deviation moving variance may be used.
- FIG. 9 is a graph showing the first variation data D5 and the second variation data D8.
- the horizontal axis of the graph is the same as the horizontal axis of the graph of FIG. 5A.
- the vertical axis of the graph indicates standard deviation.
- the first variation data D5 is data indicating the moving standard deviation of the first difference data D4 shown in FIG. 8A.
- the second variation data D8 is data indicating the moving standard deviation of the second difference data D7 shown in FIG. 8B.
- the number of frames used for the calculation of the moving standard deviation is 21 in both the first variation data D5 and the second variation data D8, but any number that can obtain a statistically significant standard deviation is available. Well, not limited to 21.
- first variation data D5 and the second variation data D8 are standard deviations, they do not include negative values. Therefore, the first variation data D5 and the second variation data D8 can be regarded as data converted so that the first difference data D4 and the second difference data D7 can be subtracted.
- the image generation unit 8 sets the data obtained by calculating the difference between the first variation data and the second variation data obtained from the same time-series pixel data as the third difference data, and M time-series data M pieces of third difference data corresponding to each of the pixel data are calculated (step S7).
- FIG. 10 is a graph showing the third difference data D9.
- the horizontal axis of the graph is the same as the horizontal axis of the graph of FIG. 5A.
- the vertical axis of the graph is standard deviation.
- the third difference data D9 is data indicating a difference between the first variation data D5 and the second variation data D8 shown in FIG.
- the image generation unit 8 outputs the M pieces of third difference data obtained in step S7 to the display control unit 12 as moving image data subjected to the processing excluding the second frequency component data and the third frequency component data.
- the moving image indicated by the moving image data is a monitoring image.
- the display control unit 12 displays the moving image indicated by the moving image data on the display unit 13.
- a monitoring image included in this moving image for example, there are an image I15 shown in FIG. 11 and an image I18 shown in FIG.
- FIG. 11 is an image diagram showing an image I13, an image I14, and an image I15 generated based on the frame at time T1.
- the image I13 is a frame image at time T1 in the moving image indicated by the M pieces of first variation data obtained in step S3 of FIG.
- the image I14 is an image of a frame at time T1 in the moving image indicated by the M pieces of second variation data obtained in step S6 of FIG.
- the difference between the image I13 and the image I14 is an image I15 (monitoring image).
- FIG. 12 is an image diagram showing an image I16, an image I17, and an image I18 generated based on the frame at time T2.
- the image I16 is an image of the frame at time T2 in the moving image indicated by the M pieces of first variation data obtained in step S3.
- the image I17 is an image of a frame at time T2 in the moving image indicated by the M pieces of second variation data obtained in step S6.
- the difference between the image I16 and the image I17 is the image I18 (monitoring image).
- Each of the images I13 to I18 shown in FIGS. 11 and 12 is an image in which the standard deviation is multiplied by 5000 times.
- the image I15 shown in FIG. 11 is an image taken before the gas is ejected from the point SP1 shown in FIG. 4, the state in which the gas is emitted from the point SP1 does not appear in the image I15.
- the image I18 shown in FIG. 12 is an image taken at the time when the gas is ejected from the point SP1, a state in which the gas is emitted from the point SP1 appears in the image I18.
- the image generation unit 8 (FIG. 1A) generates the moving image data by performing the process of removing the second frequency component data included in the moving image data D1 of the infrared image. Then, the display control unit 12 causes the display unit 13 to display the moving image (the moving image of the monitoring image) indicated by the moving image data. Therefore, according to the present embodiment, the gas leakage and the background temperature change occur in parallel, and even when the background temperature change is larger than the temperature change due to the leaked gas, the state of gas leakage is monitored. It can be displayed as an image video.
- the infrared image is configured by two-dimensionally arranging a plurality of pixels.
- the background including the monitoring target is virtually divided into a plurality of regions corresponding to the plurality of pixels.
- the pixel data of each pixel indicates the background temperature of the corresponding area.
- the background temperature of that area when there is gas in that area background temperature with gas
- the area when there is no gas in that area Background temperature background temperature without gas
- FIG. 13 is an explanatory diagram for explaining the relationship between an infrared image taken at time T10 and a background including gas.
- the gas is leaking from the monitoring target of the gas leak (for example, the place where the gas transport pipes are connected to each other) and drifts in the space.
- the infrared image is configured by two-dimensionally arranging M pixels from the first to the Mth.
- M is a plurality.
- the background is virtually divided into M areas from the first to the M-th corresponding to each of the M pixels.
- the first pixel corresponds to the first region
- the pixel data of the first pixel indicates the background temperature of the first region.
- the Jth pixel corresponds to the Jth region, and the pixel data of the Jth pixel indicates the background temperature of the Jth region.
- the background temperature with gas in the Jth region and the background temperature without gas in the Jth region are required.
- FIG. 14 is an explanatory view illustrating the relationship between an infrared image taken at time T11 different from time T10 and a background containing gas. At time T11, there is no gas in the Jth region. This is because the leaked gas fluctuates. Gas fluctuations are caused by wind. The present inventor has found that when the leaked gas fluctuates, there is a high possibility that a state in which there is gas in the Jth region and a state in which there is no gas occur in the time series.
- FIG. 15 is an image diagram showing an infrared image obtained by photographing an outdoor test place. This is an infrared image obtained by shooting a moving image with an infrared camera. There is a point SP4 at the test place where gas can be ejected.
- FIG. 16 is a graph showing a temperature change at the point SP4 of the test place.
- the vertical axis of the graph indicates the background temperature.
- the horizontal axis of the graph indicates the frame order. For example, 160 means the 160th frame.
- the frame rate is 30 fps.
- the temperature of the gas is 11.4 degrees, which is lower than the temperature of the test place (that is, the place where the infrared image was taken).
- the background temperature of the point SP4 has dropped from the 170th frame. This is because the time corresponding to this frame is the time when gas ejection is started at the point SP4. Gas continues to be ejected at point SP4. After the 170th frame, the background temperature at the point SP4 is not constant and continues to change because the jetted gas fluctuates, and there are a state where there is gas and a state where there is no gas at the point SP4. Because.
- the background temperature with gas and the background temperature without gas are obtained by using the change data of the background temperature after gas ejection (the amplitude of the graph indicating the fluctuation of the background temperature).
- a filter that transmits a wavelength range that is absorbed by the gas to be detected and a filter that does not transmit the wavelength range are prepared.
- the background temperature with gas and the background temperature without gas are obtained by utilizing the phenomenon that the leaked gas fluctuates.
- two types of filters and a mechanism for switching between them are not required.
- the background temperature with gas and the background temperature without gas are calculated by the arithmetic processing unit 9 (FIG. 1A).
- the point SP4 (FIG. 15) is one pixel, and the graph shown in FIG. 16 is time-series pixel data (FIG. 2) corresponding to the point SP4.
- the arithmetic processing unit 9 is a predetermined pixel (for example, among a plurality of (M) pixels constituting the infrared image, among backgrounds including a monitoring target (not shown)).
- the background temperature indicated by the pixel data of the predetermined pixel is set to the background temperature with gas, and when there is no gas in the region, the pixel data of the predetermined pixel
- the background temperature shown is the gas-free background temperature
- the gas-determined background temperature and the gas-free background temperature are determined using the time-series pixel data (FIG. 2) corresponding to a predetermined pixel as a temperature determination process, and red Temperature determination processing is performed for each of a plurality (M) of pixels constituting the infrared image, with each of a plurality (M) of pixels constituting the outer image as a predetermined pixel.
- the arithmetic processing unit 9 uses the time-series pixel data corresponding to the first pixel to calculate the background temperature with gas and the background temperature without gas with respect to the first region.
- the background temperature with gas and the background temperature without gas are determined for the second region, ..., corresponding to the M-1th pixel
- the gas-containing background temperature and the gas-free background temperature are determined for the M ⁇ 1th region, and the M-th pixel is determined using the time-series pixel data corresponding to the M-th pixel. For the region, determine the background temperature with gas and the background temperature without gas.
- the calculation processing unit 9 and the image processing unit 8 constitute a determination unit.
- the determination unit generates time-series pixel data from a plurality of infrared images (moving image data D1) input from the image data input unit 15, and based on each time-series pixel data of a plurality of pixels constituting the infrared image. Then, the background temperature with gas indicating the background temperature when there is a gas corresponding to each of the plurality of pixels and the background temperature without gas indicating the background temperature when there is no gas are determined.
- the calculation unit 10 uses the background temperature with gas and the background temperature without gas determined by the determination unit to calculate the concentration / thickness product of the gas corresponding to each of a plurality of (M) pixels constituting the infrared image. To do.
- FIG. 17 is an explanatory diagram for explaining this.
- a two-dimensional image sensor 6 included in the infrared camera 2 illustrated in FIG. 1A includes sensor pixels corresponding to pixels (FIGS. 13 and 14). That is, the two-dimensional image sensor 6 is configured by two-dimensionally arranging M sensor pixels from the first to the Mth.
- the two-dimensional image sensor 6 includes a Jth sensor pixel corresponding to the Jth pixel.
- the Jth sensor pixel corresponds to the Jth region.
- Igas is an expression indicating a signal (background signal with gas) output from a sensor pixel when there is gas in a region corresponding to the sensor pixel.
- Inogas is an expression indicating a signal (background signal without gas) output by the sensor pixel when there is no gas in the region.
- Igas is an equation indicating a signal output from the J-th sensor pixel when there is gas in the J-th region.
- Inogas is an expression indicating a signal output by the Jth sensor pixel when there is no gas in the Jth region.
- the gas concentration / thickness product ct is included in the equation of ⁇ gas ( ⁇ ), and this equation is included in the equation of Igas.
- the gas concentration The thickness product ct is obtained.
- the background infrared ray amount Pback corresponds to the background temperature.
- the gas concentration thickness product ct of the Jth region will be described as an example.
- the background signal with gas output from the J-th sensor pixel can be obtained from the background temperature with gas indicated by the pixel data of the J-th pixel.
- the gas temperature can be approximated to the air temperature, so it is the same as the air temperature.
- the temperature is obtained using a temperature sensor.
- Humidity is determined using a humidity sensor. Humidity has a small effect on the gas concentration thickness product, and therefore humidity may be set to 50%, for example, instead of obtaining humidity with a humidity sensor.
- the distance the distance between the infrared camera 2 and the subject set in the infrared camera 2 is used.
- the background infrared ray amount Pback of the Jth region is obtained using the Inogas equation. More specifically, as can be seen from the Inogas equation, if the background signal without gas, temperature, humidity, and the distance between the infrared camera 2 and the subject (monitoring target for gas leakage) are known, the background infrared ray amount Pback can be obtained.
- the gasless background signal output from the Jth sensor pixel can be obtained from the gasless background temperature indicated by the pixel data of the Jth pixel.
- the temperature, humidity, and distance can be obtained as described above. Since there is no formula for obtaining the background infrared ray amount Pback from these parameters (background signal without gas, temperature, humidity, distance), a table showing the relationship between these parameters and the background infrared ray amount Pback is created in advance. Using this table and parameters (and further using interpolation if necessary), the background infrared ray amount Pback is obtained. Note that the background infrared ray amount Pback may be obtained using convergence calculation without using a table.
- the gas detection image processing apparatus 3 (FIG. 1A) performs a gas concentration / thickness product calculation process using gas fluctuations to obtain an estimated value of the gas concentration / thickness product.
- FIG. 18 is a flowchart for explaining processing for obtaining an estimated value of the gas concentration thickness product.
- display control unit 12 causes display unit 13 to display a monitoring image generated using a background infrared image including a gas leak monitoring target.
- the monitor of gas leakage uses the pixel of interest in the monitor image displayed on the display unit 13 as the pixel of interest, and operates the input unit 14 to input the position of the pixel of interest (that is, Specify the pixel of interest).
- the target pixel is an example of a predetermined pixel, and is a pixel corresponding to a region where leaked gas is drifting.
- the pixel corresponding to the J-th region shown in FIGS. 13 and 14, that is, the J-th pixel can be set as the pixel of interest.
- the pixel of interest will be described by taking the Jth pixel as an example.
- the arithmetic processing unit 9 supplies gas to the J-th region corresponding to the J-th pixel in the J-th pixel (target pixel, predetermined pixel) among the M pixels. If there is a gas, the background temperature indicated by the pixel data of the Jth pixel is the background temperature with gas, and if there is no gas in the Jth region, the background temperature indicated by the pixel data of the Jth pixel is no gas. A background temperature with gas and a background temperature without gas are determined based on the time-series pixel data of the Jth pixel (FIG. 2).
- the arithmetic processing unit 9 sets a group of a predetermined number of frames that are smaller than the number of K (plural) frames shown in FIG. Among the series pixel data, the background temperature with gas and the background temperature without gas are determined from the background temperatures indicated by the pixel data included in the frame group (step S31).
- the frame group is composed of a predetermined number of frames in which the chronological order is continuous.
- the predetermined number is 41, for example.
- the frame group includes a frame of interest, 20 consecutive frames immediately before this frame, and 20 consecutive frames immediately after this frame.
- FIG. 19 is a graph showing the relationship between the temperature change at the test site point SP4 (FIG. 15) and the frame group.
- the vertical and horizontal axes of the graph and the line indicating the temperature change are the same as those shown in FIG. 5A.
- the line indicating this temperature change is time-series pixel data of the pixel corresponding to the point SP4. For example, when the frame of interest is the 200th frame, one frame group is composed of the 180th to 220th frames.
- the first frame group is constructed.
- the frame of interest is the 21st frame, and is composed of the 1st to 41st frames.
- the last frame group is a frame group including 260th to 300th frames.
- the first to last frame groups are a plurality of frame groups with different frame combinations.
- the arithmetic processing unit 9 determines the maximum value and the minimum value of the background temperature from the background temperatures indicated by the pixel data included in the frame group among the time-series pixel data of the Jth pixel.
- the frame group is the first frame group.
- the maximum value is one of the background temperature with gas or the background temperature without gas
- the minimum value is the other of the background temperature with gas or the background temperature without gas. This is determined by the relationship among the gas temperature, the background temperature with gas, and the background temperature without gas. This will be described with reference to FIGS. 20A and 20B.
- 20A and 20B are explanatory diagrams for explaining the relationship.
- the background temperature indicated by the pixel data of the Jth pixel (predetermined pixel) includes a background temperature with gas and a background temperature without gas. When the subject contains gas, there is a background temperature with gas between the gas temperature and the background temperature without gas.
- the maximum background temperature is the gas-free background temperature
- the minimum background temperature is the gas-present background temperature.
- the maximum value of the background temperature is the background temperature with gas
- the minimum value of the background temperature is the background temperature without gas. It becomes.
- the air temperature may be the gas temperature.
- the background temperature with gas and the background temperature without gas in step S31 are determined.
- the background temperature with gas and the background temperature without gas can be measured with one infrared camera 2 without requiring two types of filters and a mechanism for switching between them.
- the calculation unit 10 calculates the concentration / thickness product of the gas in the J-th region using the background temperature with gas and the background temperature without gas determined in step S31 (step S32).
- the gas concentration / thickness product is calculated using the background temperature with gas and the background temperature without gas of the first frame group.
- the calculation unit 10 compares the gas concentration / thickness product calculated in step S32 with the candidate value, and stores the larger one as the candidate value (step S33). As will be described later, the calculation unit 10 uses the finally stored candidate value as an estimated value of the concentration thickness product of the leaked gas. The initial value of the candidate value is 0. Therefore, the calculation unit 10 uses the gas concentration / thickness product calculated in step S32, that is, the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas in the first frame group as candidate values.
- the gas concentration / thickness product calculated in step S32 that is, the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas in the first frame group as candidate values.
- the arithmetic processing unit 9 determines whether or not the target frame is the last (step S34). For example, when the number of frames is 300, the 280th frame is the last. This is because if the frame of interest is the 280th frame, the 260th to 300th frames constitute a frame group (the last frame group).
- the process returns to step S31.
- the arithmetic processing unit 9 creates the next frame group by shifting the frame of interest by one in time series.
- a frame group is created when the target frame is the 22nd frame.
- This frame group is the second frame group, and is composed of the second to 42nd frames.
- the arithmetic processing unit 9 determines the background temperature with gas and the background temperature without gas from the background temperatures indicated by the pixel data included in the second frame group in the time-series pixel data of the J-th pixel ( Step S31).
- the calculation unit 10 calculates the concentration-thickness product of the gas in the Jth background using the maximum background temperature value and the minimum background temperature value determined in step S31 (step S32).
- the gas concentration / thickness product is calculated using the background temperature with gas and the background temperature without gas in the second frame group.
- the calculation unit 10 compares the gas concentration / thickness product calculated in step S32 with the candidate value, and stores the larger one as the candidate value.
- the calculation processing unit 9 and the calculation unit 10 repeat the processing from step S31 to step S33 until it is determined that the target frame is the last (Yes in step S34). That is, the arithmetic processing unit 9 prepares a plurality of frame groups with different frame combinations, and determines the background temperature with gas and the background temperature without gas for each of the plurality of frame groups. The calculation unit 10 calculates the gas concentration / thickness product for each of the plurality of frame groups using the background temperature with gas and the background temperature without gas determined by the determination unit included in the calculation unit 10.
- the calculation unit 10 estimates the candidate value stored in step S33 as the concentration thickness product of the drifting gas. (Step S35). In this way, the calculation unit 10 sets the maximum value among the gas concentration / thickness products of each of the plurality of frame groups calculated in step S32 as the estimated value of the drifting gas concentration / thickness product.
- FIG. 21 is an image diagram showing a transition of an image displayed on the display unit 13 (FIG. 1A) when the process of FIG. 18 is executed. These are infrared images obtained by photographing the moving image of the test place described in FIG.
- the part in the frame indicated by the dotted line is an image obtained by visualizing the gas concentration / thickness product calculated in step S32.
- This is composed of a pixel of interest (Jth pixel) and pixels located around it. These pixels correspond to values obtained by multiplying the gas concentration / thickness product calculated in step S32 by 100. For these pixels, the processing from step S31 to step S35 is performed.
- the gas concentration is indicated by LEL (Lower Explosive Limit).
- LEL Lower Explosive Limit
- the lower explosion limit is the lowest concentration at which a combustible gas mixed with air will cause an explosion upon ignition. 100% LEL means that the lower explosion limit has been reached. In the case of methane, when the concentration reaches 5%, it becomes 100% LEL.
- the gas concentration thickness product is indicated by LELm. m is the distance in the depth direction.
- step S32 indicates the gas concentration / thickness product calculated in step S32
- step S33 indicates the candidate value of step S33 (the maximum value of the gas concentration / thickness product calculated so far).
- step S35 the processing from step S31 to step S35 is performed for all pixels in the frame portion indicated by the dotted line
- ct and ct max are values for the target pixel.
- the image shown at time T20 is an image immediately after the start of gas ejection.
- ct represents the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas of the frame group in which the frame of interest is the frame at time T20.
- the image shown at time T21 is an image when 2 seconds have elapsed from time T20.
- ct represents the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas of the frame group in which the frame of interest is the frame at time T21.
- the concentration thickness product of the gas drifting in the region corresponding to the pixel of interest is relatively low. This means that there is relatively little gas drifting in this area.
- the image shown at time T22 is an image when 5 seconds have elapsed from time T20.
- ct represents the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas of the frame group in which the target frame is the frame at time T22.
- the concentration thickness product of the gas drifting in the region corresponding to the target pixel is relatively high. This means that there is a relatively large amount of gas drifting in this area.
- the image shown at time T23 is an image when 8 seconds have elapsed from time T20.
- ct represents the gas concentration / thickness product calculated using the background temperature with gas and the background temperature without gas of the frame group in which the target frame is the frame at time T23.
- the concentration thickness product of the gas drifting in the region corresponding to the target pixel is relatively low. This means that there is relatively little gas drifting in this area.
- the gas concentration / thickness product (3.5% LELm) at time T22 is an estimated value of the gas concentration / thickness product in the region corresponding to the target pixel.
- the exact value of the gas concentration thickness product in this region was 3% LELm.
- the background temperature with gas and the background temperature without gas are determined using the phenomenon that the leaked gas fluctuates. Since the fluctuation of the gas is generated by wind or the like, the temperature difference between the background temperature with gas and the background temperature without gas varies along the time axis, and as a result, the concentration thickness product also varies along the time axis.
- the maximum value of the concentration / thickness product is regarded as the estimated value of the concentration / thickness product, and this estimated value is regarded as the concentration / thickness product.
- the estimated value of the gas concentration / thickness product can fall within the range of 0.5 to 2 times the exact value of the gas concentration / thickness product.
- the reason why the predetermined number of frames is 41 frames will be described.
- the gas concentration / thickness product of the Jth region (region corresponding to the target pixel) shown in FIGS. 13 and 14 is calculated, and the maximum value among them is calculated as the gas value. This is the estimated value of the concentration-thickness product. If there is a gas in the Jth region or no gas in the Jth region over the entire period of a frame group, the gas concentration / thickness product in that frame group cannot be calculated. In order to calculate the gas concentration / thickness product, it is necessary to generate a state in which there is gas in the Jth region and a state in which there is no gas during the period of one frame group.
- the period of one frame group is lengthened, it is possible to reliably generate a state where there is gas in the Jth region and a state where there is no gas.
- the clouds move and block sunlight, or when clouds blocking the sunlight move, the background temperature changes. If the period of one frame group is too long, the possibility of being affected by this increases. On the other hand, if the period of one frame group is made too short, both the state where there is gas in the Jth region and the state where there is no gas hardly occur.
- the period of one frame group is about 1.4 seconds from these viewpoints.
- the predetermined number of frames is 41 frames. If the frame rate changes, the number of the predetermined number of frames changes. Depending on the assumed conditions (for example, wind speed), it is not always necessary to be 1.4 seconds, and may be changed.
- FIG. 22 is an image diagram showing three infrared images selected from the moving image data D1 of the infrared image.
- the image I30 is an infrared image of the test place taken at time T30.
- the image I31 is an infrared image of the test place taken at time T31 one second after time T30.
- the image I32 is an infrared image of the test place taken 2 seconds after the time T30.
- gas is ejected at the point SP5.
- an image of a moving object (running train) is shown from the left to the center in the upper part of the image I31.
- an image of a moving object is shown from left to right.
- the image generation unit 8 performs image processing on the moving image data D1 of the infrared image to generate moving image data of the monitoring image (step S100 in FIG. 3).
- the display control unit 12 causes the display unit 13 to display the moving image of the monitoring image indicated by the moving image data of the monitoring image.
- examples of the monitoring image are an image I15 shown in FIG. 11 and an image I18 shown in FIG.
- the monitoring image includes a gas image in which the leaked gas is visualized.
- the monitoring image includes the image of the moving object.
- both the gas image and the moving object image are shown as white images.
- the density-thickness product visualized image by using the density-thickness product visualized image, it is possible to identify the pixels constituting the gas image included in the monitoring image and the pixels constituting the moving object image included in the monitoring image. To do.
- the density-thickness product visualized image is an image that has been processed to visualize the concentration-thickness product of gas.
- the calculation unit 10 performs a process of calculating a density thickness product for each infrared image constituting the moving image data D1 used in the process of step S100 (step S101 in FIG. 3).
- the density-thickness product is calculated for regions corresponding to each of a plurality of pixels (in other words, all pixels) constituting the infrared image. More specifically, for example, with respect to the infrared image at time T10 shown in FIG. 13, the density-thickness product is calculated for M regions corresponding to the M pixels.
- the calculation unit 10 uses the infrared image to calculate the density thickness product corresponding to each of the plurality of pixels constituting the infrared image as the identification value.
- An example of a method for calculating the concentration / thickness product is illustrated in FIG.
- the image generation unit 8 generates an image (density thickness product visible image) in which the density thickness product calculated in step S101 in FIG. 3 is visualized.
- FIG. 23 is an image diagram showing various images generated using the density-thickness product.
- the image I31a is an image (density / thickness product visualized image) obtained by visualizing a value obtained by multiplying a density / thickness product corresponding to each of a plurality of pixels constituting the image I31 illustrated in FIG.
- the density / thickness product visualized image is an image obtained by visualizing the density / thickness product calculated on the assumption that gas exists in a region corresponding to the entire surface of the image, and is not an image showing a gas image.
- the density-thickness product visualized image is the pixel constituting the gas image included in the monitoring image and the moving object included in the monitoring image. It will contain the pixels that make up the image.
- the gas concentration is indicated by LEL (lower explosion limit). 100% LEL means that the lower explosion limit has been reached. In the case of methane, a concentration of 5% is 100% LEL.
- the concentration thickness product is indicated by LELm. m is the distance in the depth direction.
- the image I31a has 256 gradations, and an area of 2.55% LELm or more is shown in white.
- the concentration thickness product (for example, a value of 200% LELm or more) that greatly exceeds the lower limit of gas explosion is an abnormal value.
- the pixel is not a pixel constituting the gas image included in the monitoring image, but an image of a moving object included in the monitoring image. It can be regarded as a constituent pixel.
- an image I31b shown in FIG. 23 is an image obtained by visualizing a value obtained by multiplying a density thickness product corresponding to each of a plurality of pixels constituting the image I31 shown in FIG.
- an area of 200% LELm or more is also clipped and calculated as 200. Therefore, in the case of 256 gradation display, an area of 200% LELm or more is displayed as the same gradation value (200).
- the identification unit 11 compares the density / thickness product corresponding to each of a plurality of pixels (all pixels) constituting the density / thickness product visualized image with a predetermined threshold (for example, 200% LELm). Then, the pixel corresponding to the density / thickness product exceeding the threshold is specified as the pixel constituting the moving object image included in the monitoring image. Thereby, the identification part 11 can identify the pixel which comprises the gas image contained in the monitoring image, and the pixel which comprises the image of moving bodies other than the gas image contained in the monitoring image (step S102).
- a predetermined threshold for example, 200% LELm
- the identification unit 11 compares the absolute value of the identification value corresponding to each of the plurality of pixels constituting the infrared image with a predetermined threshold value, and the identification value exceeding the threshold value.
- a pixel corresponding to the absolute value of is specified as a pixel constituting a non-gas image (for example, an image of a moving object other than a gas image).
- the image generation unit 8 generates a visualized image that visualizes an image (in other words, a pseudo image of a moving object image) formed by pixels constituting the moving object image included in the monitoring image specified by the identification unit 11. .
- an image I31c shown in FIG. 23 is an image (visualized image) obtained by binarizing the density / thickness product corresponding to each of a plurality of pixels constituting the image I31a with the threshold value. Pixels corresponding to density / thickness products exceeding the threshold are shown in white, and pixels corresponding to density / thickness products below the threshold are shown in black.
- the display control unit 12 wants to display the gas region on the display unit 13, among the pixels of the monitoring image generated by the image generation unit 8, the region indicated by white in the image 131 a in FIG. 23 is not displayed. (Make the area black). As a result, an image excluding the non-gas region should be displayed on the display unit 13.
- the calculation unit 10 When the calculation unit 10 obtains an estimated value of the concentration / thickness product according to the flowchart shown in FIG. 18 described in [Gas concentration / thickness product calculation process], the calculation unit 10 forms an infrared image. An estimated value corresponding to each of a plurality (M) of pixels is calculated. The identification unit 11 compares an estimated value corresponding to each of a plurality of (M) pixels constituting the infrared image with a threshold value as a concentration thickness product of a gas serving as an identification value.
- the arithmetic processing unit 9 may perform the same processing by calculating a value correlated with the concentration-thickness product as described in the second mode and comparing it with the threshold value.
- the first aspect of the present embodiment generates a visualized image such as an image I31c by using a density / thickness product visualized image such as an image I31a, and a gas leak monitor can An image and a moving object image can be distinguished.
- the gas value monitor can identify the gas image and the moving object image using the correlation value visualized image.
- the correlation value visualized image is an image obtained by visualizing the correlation value of the gas concentration thickness product.
- the correlation value is a value obtained by dividing the temperature difference between the background temperature with gas and the background temperature without gas by the temperature difference between the background temperature without gas and the gas temperature.
- the former temperature difference can be rephrased as the amplitude of the temperature change caused by the gas.
- An example of a method for determining the background temperature with gas and the background temperature without gas is described in [Calculation process of gas concentration / thickness product] in the first aspect of the present embodiment.
- the temperature difference between the background temperature without gas and the background temperature without gas is the amplitude that is the difference between the maximum value and the minimum value of the temperature change, and the temperature between the gas temperature and the background temperature without gas.
- the difference can be approximated by a temperature difference between the gas temperature and the average value of the maximum value and the minimum value of the temperature change.
- the gas detection image processing apparatus 3 can calculate the correlation value only by finding the maximum value and the minimum value of the temperature change without determining the background temperature with gas and the background temperature without gas.
- FIG. 24 is a flowchart illustrating the second aspect of the present embodiment.
- the second aspect is included in the monitoring image generation process (step S200), the correlation value calculation process of the gas concentration / thickness product (step S201), and the pixels and the monitoring image constituting the gas image included in the monitoring image. It is constituted by a discrimination process (step S202) with pixels constituting an image of a moving body other than a gas image.
- the monitoring image generation process (step S200) is the same as the monitoring image generation process (step S100) of the first aspect shown in FIG.
- the correlation value calculation process (step S201) will be described.
- a correlation value is used instead of the concentration thickness product.
- FIG. 25 is an explanatory diagram for explaining the basic characteristics of the concentration-thickness product.
- the concentration-thickness product has three basic characteristics (1) to (3).
- the concentration thickness product When the concentration thickness product is the same, the gas image appears darker in the monitoring image when the difference between the gas-free background temperature and the gas temperature is larger. That is, the amplitude of the temperature change increases.
- the gas-free background temperature When the gas-free background temperature is the same and the gas temperature is the same, the gas image appears darker in the monitoring image when the concentration thickness product is larger. That is, the amplitude of the temperature change increases.
- the background temperature with gas is higher than the gas temperature and lower than the background temperature without gas.
- the background temperature with gas is lower than the gas temperature and higher than the background temperature without gas.
- the correlation value is the main factor that determines the size of the concentration-thickness product. If the correlation value increases, the concentration-thickness product increases. If the correlation value decreases, the concentration-thickness product increases. Becomes smaller.
- the density-thickness product visualized image (for example, the image I31a shown in FIG. 23) is calculated on the assumption that there is gas in a region corresponding to each of a plurality of pixels constituting this image (region corresponding to the entire surface of the image).
- the This causes the following two problems ⁇ 1> and ⁇ 2>.
- FIG. 26 is an explanatory diagram for explaining the problem ⁇ 1> of the density / thickness product visualized image.
- FIG. 27 is an explanatory diagram for explaining the problem ⁇ 2> of the density-thickness product visualized image.
- the gas temperature is one of the gas-free background temperature or the gas-containing background temperature and the gas-free background temperature or the gas-containing background temperature. There is no time between them.
- the density thickness product corresponding to this pixel cannot be calculated for the pixel having such a temperature relationship.
- the correlation value is defined by the above (formula)
- Correlation value is determined. Accordingly, in the correlation value visualized image, since all correlation values corresponding to each of the plurality of pixels constituting this image are determined, inconvenience occurs when performing image processing using all of the plurality of pixels constituting this image. Absent.
- the correlation value visualized image is an approximate image of the density-thickness product visualized image. According to the correlation value visualized image, the problem ⁇ 1> can be eliminated as described above. Problem ⁇ 2> also occurs in the case of a correlation value.
- FIGS. 29A and 29B in the case of the correlation value, the calculation unit 10 (FIG. 1A), when the temperature difference between the gas temperature and the gas-free background temperature is equal to or less than a predetermined threshold Th, The correlation value is calculated by correcting the difference.
- FIG. 29A is a first explanatory diagram illustrating correction of a temperature difference between the gas temperature and the gas-free background temperature in the correlation value.
- FIG. 29B is a second explanatory diagram illustrating correction of the temperature difference between the gas temperature and the no-gas background temperature in the correlation value.
- the correction is clip processing.
- a temperature difference for example, 5 ° C.
- the correction of the temperature difference between the gas temperature and the gas-free background temperature means that when the temperature difference between the gas temperature and the gas-free background temperature is equal to or less than the threshold value Th, the threshold value Th is set as the gas temperature and the gas-free background temperature. (When the temperature difference is 3 ° C., for example, the threshold value 5 ° C.
- this temperature difference is the temperature difference.
- this temperature difference Is the temperature difference between the gas temperature and the background temperature without gas (for example, when the temperature difference is 10 ° C., 10 ° C. is the temperature difference).
- FIG. 30 is an image diagram showing various images of the test place taken in the daytime.
- FIG. 31 is an image diagram showing various images of a test place taken at night.
- an image I40 and an image I50 are infrared images of a test place.
- the image I41 is a density / thickness product visualized image generated using the image I40.
- the image I51 is a density / thickness product visualized image generated using the image I50. In the image I41 and the image I51, the density thickness product increases as the density becomes lighter, and the density thickness product decreases as the density becomes darker.
- the density thickness product corresponding to the pixels constituting the white portion is large.
- the density-thickness product is large depending on whether it is really large or large due to an error.
- the error of the density thickness product becomes large. For this reason, as shown in the image I51, even if the area of the white portion is wide, it is not known whether the density thickness product is really large or large due to an error.
- the image I43 is a correlation value visualized image generated using the image I40.
- the image I53 is a correlation value visualized image generated using the image I50.
- the threshold value Th (FIGS. 29A and 29B) is set to 5 ° C., the “temperature difference between the gas temperature and the background temperature without gas” is corrected, and the correlation value is calculated. A portion having a large correlation value changes from a white portion to a black portion or a gray portion due to an error, and a portion having a really large correlation value is indicated by a white portion.
- the image I44 is a correlation value visualized image generated using the image I40.
- the image I44 is an image for comparison with the image I43, and the correlation value is calculated with the threshold Th set to 0.1 ° C.
- the threshold value Th0.1 ° C. is set for convenience in order to prevent the correlation value from becoming infinite, and indicates that correction is not performed substantially.
- the image I54 is a correlation value visualized image generated using the image I50.
- the image I54 is an image to be compared with the image I53, and the correlation value is calculated by setting the threshold Th to 0.1 ° C.
- the difference between the gas temperature and the gas-free background temperature is small (for example, at night), the error of the correlation value becomes large. Therefore, in the image I54 as well as the image I51, the area of the white portion is widened and the concentration thickness product is appropriately It turns out that it becomes impossible to calculate.
- calculation unit 10 performs a process of calculating a correlation value for each infrared image constituting moving image data D1 used in the process of step S200 (step S100).
- a correlation value is calculated for a region corresponding to each of a plurality of pixels (in other words, all pixels) constituting the infrared image. More specifically, for example, with respect to the infrared image at time T10 shown in FIG. 13, correlation values are calculated for M regions corresponding to the M pixels. That is, the calculation unit 10 calculates the correlation value corresponding to each of the plurality of pixels constituting the infrared image as the identification value using the infrared image.
- the arithmetic processing unit 9 is an area corresponding to a predetermined pixel among a plurality (M) of pixels constituting an infrared image in a background (for example, FIGS. 13 and 14) including a monitoring target (not shown).
- a background for example, FIGS. 13 and 14
- the background temperature indicated by the pixel data of the predetermined pixel is the background temperature with gas
- the background temperature indicated by the pixel data of the predetermined pixel is the background temperature without gas.
- the background temperature with gas and the background temperature without gas are determined for each of a plurality (M) of pixels constituting the infrared image, with each of a plurality (M) of pixels constituting the infrared image as a predetermined pixel. To do.
- the calculation unit 10 uses the background temperature with gas and the background temperature without gas determined by the arithmetic processing unit 9 and the above (formula), and uses a plurality of (M) pixels constituting an infrared image as an identification value. Correlation values corresponding to each of these are calculated.
- the image generation unit 8 generates an image (correlation value visualized image) obtained by visualizing the correlation value calculated by the calculation unit 10.
- the identification unit 11 calculates a spatial change amount (for example, an edge amount) of the identification value for each of a plurality of pixels constituting the correlation value visualized image, and based on the calculated spatial change amount, other than the gas image
- a spatial change amount for example, an edge amount
- the spatial change amount is a value obtained by spatially differentiating each identification value of a plurality of pixels constituting the correlation value visualized image.
- the identification unit 11 uses a value obtained by spatially differentiating the identification value as a differential value, calculates a differential value of each of the plurality of pixels constituting the correlation value visualized image, and compares the calculated differential value with a predetermined threshold value. Then, the pixel corresponding to the differential value exceeding the threshold is specified as the pixel constituting the moving object image other than the gas image.
- the identification unit 11 uses the correlation value visualized image, the pixels constituting the gas image included in the monitoring image, and the pixels constituting the moving object other than the gas image included in the monitoring image. Is identified (step S202 in FIG. 24). This process will be described in detail with reference to FIGS. 32 and 33.
- FIG. FIG. 32 is a flowchart for explaining the identification processing (step S202) shown in FIG. This process includes an edge amount calculation process (step S202-1) of a correlation value visualized image, a correction amount calculation process (step S202-2), and a monitoring image correction process (step S202-3).
- FIG. 33 is an image diagram showing various images related in the identification processing (step S202) shown in FIG.
- the image I43 is the correlation value visualized image described with reference to FIG. 30 and includes pixels constituting a gas image included in the monitoring image and pixels constituting a moving object other than the gas image.
- the correlation value visualized image is an image in which the correlation value is visualized, and is not an image showing a gas image or an image of a moving object other than the gas image.
- the correlation value corresponding to the pixels constituting the gas image included in the monitoring image has a gradual spatial change. In other words, the difference between the correlation value corresponding to the target pixel and the correlation value corresponding to the pixels located around this pixel is relatively small. This is because the leaked gas drifts slowly while fluctuating.
- the target pixel can be determined not as a pixel constituting the gas image included in the monitoring image but as a pixel constituting the moving object image included in the monitoring image.
- a pixel of interest that has a relatively large difference can be identified by an edge amount.
- the edge amount is a value related to the difference between the pixel value of the pixel of interest and the pixel values of the pixels located around this pixel.
- the edge amount can be calculated, for example, by adding the absolute value of the difference from the surrounding 24 pixels (the neighborhood of 5 ⁇ 5 pixels).
- the edge amount can also be calculated by a Sobel filter or the like.
- the identification unit 11 calculates an edge amount for each of a plurality of pixels constituting the correlation value visualized image, and based on the calculated edge amount (that is, the spatial change of the correlation value is calculated). Based on this, a pixel constituting the gas image included in the monitoring image and a pixel constituting the moving body image included in the monitoring image are identified.
- the identification unit 11 calculates an edge amount corresponding to each of a plurality of pixels constituting the correlation value visualized image as shown in the image I43 (step S202-1).
- the image I60 is an image in which the edge amount corresponding to each of the plurality of pixels constituting the image I43 is visualized.
- the identification unit 11 performs a process of calculating the correction amount (step S202-2). More specifically, a pixel whose edge amount exceeds a predetermined threshold value is not a pixel that constitutes a gas image but a pixel that constitutes an image of a moving object other than the gas image.
- the identification unit 11 uses a value obtained by multiplying a value obtained by subtracting the edge amount by a constant value for a pixel whose edge amount exceeds the threshold value as a correction value, and divides the value indicated by the pixel by the correction value. On the other hand, correction is not performed for pixels whose edge amount is equal to or smaller than the threshold value. In this manner, the identification unit 11 corrects the identification value of the pixels constituting the moving object image (non-gas image) other than the gas image included in the monitoring image specified by the identification unit 11.
- the image I61 is an image in which a correction amount corresponding to each of a plurality of pixels constituting the image I60 is visualized.
- the correction amount increases as it becomes whitish, and the correction amount decreases as it becomes darker.
- the image generation unit 8 generates an image obtained by correcting the monitoring image with the correction amount calculated in step S202-2 (step S202-3). For example, a pixel value corresponding to each of a plurality of pixels constituting the monitoring image is divided by the correction amount of each pixel.
- the image I70 is a monitoring image before correction generated using the image I40 shown in FIG.
- the image I71 is an image obtained by correcting the image I70 with the correction amount calculated in step S202-2.
- the moving body image changes from a white portion to a black portion or a gray portion because the correction amount is large.
- the gas image is an uncorrected amount, it remains a white portion.
- the white portion located at the center indicates a gas image.
- there are many white portions other than the gas image but in the image I71, these white portions are weakened.
- the image I71 that is the corrected monitoring image can suppress the luminance of the moving object image compared to the luminance of the gas image.
- the edge amount calculation process (step S202-1), the correction amount calculation process (step S202-2), and the monitoring image correction process (step S202-3) can be expressed as follows.
- the identification unit 11 calculates an edge amount for each of a plurality of pixels constituting the correlation value visualized image, compares the calculated edge amount with a predetermined threshold value, and sets the edge amount pixel exceeding the threshold value. Identifies a pixel that constitutes an image of a moving object included in the monitoring image.
- the image generation unit 8 corrects a pixel that matches a pixel that constitutes an image of a moving object included in the monitoring image, which is specified by the identification unit 11, among a plurality of pixels that constitute the monitoring image.
- a monitoring image in which the luminance of the moving object image is suppressed compared to the luminance is generated.
- the identification unit 11 compares the absolute value of the identification value corresponding to each of the plurality of pixels constituting the infrared image with a predetermined threshold value, and the identification value exceeding the threshold value.
- a pixel corresponding to the absolute value of is specified as a pixel constituting a non-gas image (for example, an image of a moving object other than a gas image).
- the target pixel is the gas image.
- the pixel constituting the gas image and the pixel constituting the moving image are identified as pixels constituting the moving body image other than the above. That is, in the second aspect of the present embodiment, the identification is performed using a spatial change in the correlation value. On the other hand, in the third aspect of the present embodiment, the above identification is performed using a temporal change in the correlation value.
- FIG. 34 is a flowchart for explaining the third aspect of the present embodiment.
- the third mode is included in the monitoring image generation process (step S300), the correlation value calculation process of the gas concentration thickness product (step S301), the time series correlation value data calculation process (step S303), and the monitoring image. And a pixel constituting a gas image and a pixel constituting a moving body image other than the gas image included in the monitoring image (step S304).
- the monitoring image generation process (step S300) and the correlation value calculation process (step S301) are the monitoring image generation process (step S200) and the correlation value calculation process (step S200) of the second aspect shown in FIG. Since this is the same as S201), the description is omitted.
- the image generation unit 8 generates correlation value visualized images arranged in time series.
- FIG. 35 is an image diagram showing correlation value visualized images arranged in time series.
- Image I80 is an infrared image of the test location.
- the test place includes a point SP6 where gas is ejected and a point SP7 where a moving object appears. Each of the points SP6 and SP7 is indicated by one pixel.
- Images I81 to I84 illustrate some (four) correlation value visualized images in the correlation value visualized image generated by using the moving image of the infrared image of the test place.
- the image I82 is an image after 5 seconds from the image I81
- the image I83 is an image after 5 seconds from the image I82
- the image I84 is an image after 5 seconds from the image I83.
- an image formed by pixels constituting the moving object image in other words, an image corresponding to the moving object image
- Time series correlation value data (time series identification value data) calculation processing (step S303) will be described.
- FIG. 36 is an explanatory diagram for explaining time-series correlation value data (time-series identification value data).
- K be the number of correlation value visualized images arranged in time series.
- One correlation value visualized image has a correlation value corresponding to each of M (plural) pixels, that is, a correlation value corresponding to the first pixel, a correlation value corresponding to the second pixel,... It is assumed that a correlation value corresponding to the ⁇ 1st pixel and a correlation value corresponding to the Mth pixel are included.
- a correlation value corresponding to pixels at the same spatial position means a correlation value corresponding to pixels in the same order.
- the correlation value corresponding to the first pixel included in the first correlation value visualized image the correlation value corresponding to the first pixel included in the second correlation value visualized image, ..., the correlation value corresponding to the first pixel included in the K-1th correlation value visualized image, the correlation value corresponding to the first pixel included in the Kth correlation value visualized image, in time series
- the arranged data becomes the time-series correlation value data of the correlation value corresponding to the first pixel.
- the number of time-series correlation value data is the same as the number of pixels (M) constituting one correlation value visualized image.
- the calculation unit 10 calculates M (plural) time-series correlation value data, that is, time-series correlation value data of the correlation value corresponding to the first pixel, and time-series correlation value data of the correlation value corresponding to the second pixel. ..., time-series correlation value data of correlation values corresponding to the (M-1) th pixel, and time-series correlation value data of correlation values corresponding to the M-th pixel are generated.
- the identification process (step S304) between the pixels constituting the gas image included in the monitoring image and the pixels constituting the moving body image other than the gas image included in the monitoring image will be described.
- the identification is performed using a temporal change in the correlation value.
- the gas leaked from the monitored object slowly spreads while fluctuating.
- the correlation value corresponding to the pixels constituting the gas image gradually changes (slowly changes) on the time axis.
- the pixel corresponding to the correlation value can be regarded as a pixel constituting an image of a moving object other than the gas image. Therefore, the identification unit 11 determines whether the correlation value changes abruptly on the time axis by using time-series correlation value data. This will be described in detail below.
- the identification unit 11 includes, for a predetermined pixel among a plurality of (M) pixels constituting the correlation value visualized image, an abnormality in the time-series correlation value data corresponding to the predetermined pixel. Determining whether or not to include a correlation value of a different value is determined as an abnormal value determination, and a plurality of (M) pixels constituting the correlation value visualized image are used as predetermined pixels, and a plurality of the correlation value visualized images are configured. An abnormal value is determined for each of the (M) pixels, and a pixel corresponding to time-series correlation value data including a correlation value of an abnormal value is specified as a pixel constituting a moving object image included in the monitoring image. .
- the calculation unit 10 determines whether or not an abnormal correlation value is included for each of the M-th pixel time-series correlation value data from the first-pixel time-series correlation value data. For example, if the time-series correlation value data of the first pixel includes an abnormal correlation value, the calculation unit 10 configures the first pixel of the correlation value visualized image as a moving object image. It is specified as a pixel. Whether or not the correlation value is an abnormal value can be determined by a threshold value. That is, the calculation unit 10 determines that the correlation value included in the time series correlation value data exceeds the threshold value as an abnormal value.
- the identification unit 11 determines whether or not an abnormal correlation value is included in the time-series correlation value data using a histogram.
- the identification unit 11 generates a histogram indicating the appearance frequency of correlation values having the same value for each of the first to Mth time-series correlation value data shown in FIG. That is, the identification unit 11 generates a histogram of correlation values corresponding to the first pixel based on the time-series correlation value data of the correlation values corresponding to the first pixel, and the correlation corresponding to the second pixel.
- a histogram of correlation values corresponding to the second pixel is generated,..., Based on the time-series correlation value data of the correlation values corresponding to the M ⁇ 1th pixel.
- a correlation value histogram corresponding to the M ⁇ 1th pixel is generated, and the correlation value corresponding to the Mth pixel is calculated based on the time-series correlation value data of the correlation value corresponding to the Mth pixel.
- FIG. 37 is an explanatory diagram illustrating a histogram generated using time-series correlation value data.
- the histogram 20 is a histogram created based on time-series correlation value data corresponding to the point SP6 (pixel) where gas is ejected.
- the histogram 21 is a histogram created based on time-series correlation value data corresponding to the spot SP7 (pixel) where a moving object appears.
- the vertical axes of the histogram 20 and the histogram 21 indicate the appearance frequency of the same correlation value, and the horizontal axis indicates the correlation value.
- the actual correlation value is a value of 1 or less, in order to indicate the correlation value as an integer, the actual correlation value is multiplied by 40 and the value obtained by rounding down the decimal point is used as the correlation value. For the correlation value 0, a bar graph is also shown.
- the correlation value corresponding to the point SP6 (pixel) where gas is ejected fluctuates within a predetermined range (here, within a range from 0 to 2).
- the correlation value corresponding to the spot SP7 (pixel) where the moving object appears has a value (here, 17 to 20) that is different from the predetermined range (here, within the range of 0 to 2).
- the identification unit 11 determines such a correlation value as an abnormal value. That is, the identification unit 11 determines, in the histogram, a correlation value that does not belong to a group of correlation values within a predetermined range as an abnormal value.
- the predetermined range is a fluctuation range of the correlation value caused by the gas leaked from the monitoring target, and the identification unit 11 stores data of the fluctuation range in advance.
- the identification unit 11 determines a correlation value that does not belong to a group of correlation values within a predetermined range as an abnormal value in each of the first to Mth time-series correlation value data shown in FIG.
- the image generation unit 8 generates an image obtained by binarizing a pixel whose correlation value is determined to be an abnormal value and a pixel whose correlation value is determined not to be an abnormal value based on the determination by the identification unit 11.
- An example of this image is an image I90 shown in FIG.
- White pixels are pixels for which the correlation value is determined to be abnormal, and black pixels are pixels for which the correlation value is determined not to be abnormal.
- a white pixel (a pixel for which the correlation value is determined to be an abnormal value) is an image (a pseudo image of the moving object image) formed by the pixels constituting the moving object image included in the monitoring image.
- the identification unit 11 uses the gas histogram to calculate time-series correlation value data (time-series identification value data) based on the histogram (based on the temporal change in the identification value of the same value). Pixels constituting an image of a moving object other than the above are specified.
- FIG. 38 is a flowchart illustrating the fourth aspect of the present embodiment.
- the monitoring image generation process step S400
- the correlation value calculation process of the gas concentration / thickness product step S401
- the time series correlation value data calculation process step S403
- the maximum correlation value step S404
- differential image generation processing step S405
- pixels constituting a gas image included in the monitoring image and pixels constituting a moving image other than the gas image included in the monitoring image Identification processing step S406.
- the monitoring image generation process (step S400), the correlation value calculation process of the gas concentration thickness product (step S401), and the time series correlation value data calculation process (step S403) are the monitoring image of the third aspect shown in FIG. This is the same as the generation process (step S300), the correlation value calculation process (step S301), and the time series correlation value data calculation process (step S303).
- the description starts from the search for the maximum and minimum correlation values (step S404).
- the identification unit 11 (FIG. 1A) has the function of a search unit. Referring to FIG. 36, the search unit, for a predetermined pixel among a plurality of (M) pixels constituting the correlation value visualized image, the correlation value included in the time-series correlation value data corresponding to the predetermined pixel. Searching the maximum value and the minimum value is the maximum / minimum value search process, and each of a plurality of (M) pixels constituting the correlation value visualized image is a predetermined pixel, and a plurality of pixels constituting the correlation value visualized image are searched. A maximum / minimum value search process is performed for each.
- the search unit searches for the maximum value and the minimum value of the correlation value in each of the time-series correlation value data of the 1st to M-th pixels (correlation value included in the time-series correlation value data of the 1st pixel). Is searched for the maximum value and the minimum value of the correlation value included in the time-series correlation value data of the second pixel,..., The time-series correlation value of the M ⁇ 1th pixel The maximum and minimum correlation values included in the data are searched, and the maximum and minimum correlation values included in the time-series correlation value data of the Mth pixel are searched).
- the moving image of the correlation value visualized image shown in FIG. 35 is a moving image of 20 seconds, and the time of the time-series correlation value data is 20 seconds.
- the image generation unit 8 generates a maximum value visualization image that is an image obtained by visualizing the maximum value searched by the search unit and a minimum value visualization image that is an image obtained by visualizing the minimum value searched by the search unit. Then, the image generation unit 8 generates a difference image between the generated maximum value visualized image and minimum value visualized image (step S405).
- FIG. 39 is an image diagram showing various images related in the difference image generation process.
- the image I100 is a maximum value visualized image
- the image I101 is a minimum value visualized image
- the image I102 is a difference image.
- the maximum value visualized image is an image in which the maximum correlation value is visualized in each of the time-series correlation value data of the first to Mth pixels, that is, the time-series correlation of the first pixel.
- Maximum value of correlation value included in value data, maximum value of correlation value included in time-series correlation value data of second pixel,..., Correlation included in time-series correlation value data of M ⁇ 1th pixel It is the image which visualized the maximum value of the value and the maximum value of the correlation value contained in the time series correlation value data of the Mth pixel.
- the minimum value visualized image is an image in which the minimum value of the correlation value is visualized in each of the time series correlation value data of the first to Mth pixels, that is, the correlation value included in the time series correlation value data of the first pixel.
- the maximum correlation value included in the time-series correlation value data of a pixel can be regarded as the correlation value at the moment when there is the most gas in the region corresponding to the pixel, and the minimum value corresponds to the pixel. It can be regarded as the correlation value at the moment when there is the least amount of gas in the region.
- the J-th pixel will be described.
- the maximum value is regarded as the correlation value at the moment when the gas is the most in the region corresponding to the J-th pixel (J-th region).
- the minimum value can be regarded as the correlation value at the moment when the gas is the least in the region corresponding to the Jth pixel (Jth region).
- the pixel whose correlation value has a gradual spatial change among the multiple pixels forming the maximum value visualized image forms the gas image included in the monitored image It will match the pixel to be.
- the difference image it is possible to reduce the influence of the edge noise caused by the blur of the image to be monitored (including a minute image at the subpixel level).
- the identification unit 11 uses the difference image to identify a pixel constituting a gas image included in the monitoring image and a pixel constituting a moving body image other than the gas image included in the monitoring image (step) S406). This process will be described with reference to FIG. FIG. 40 is an image diagram showing various images related in the identification processing (step S406) shown in FIG.
- the identification unit 11 calculates an edge amount corresponding to each of a plurality of pixels constituting the image I102 (difference image). This is the same as step S202-1 in FIG.
- the image I103 is an image (edge amount visualized image) in which the edge amount corresponding to each of a plurality of pixels constituting the image I102 (difference image) is visualized.
- Pixels whose edge amount exceeds a predetermined threshold are pixels that constitute an image of a moving object other than the gas image included in the monitoring image, not the pixels that configure the gas image included in the monitoring image.
- the identification unit 11 compares each of the plurality of pixels constituting the image I103 (edge amount visualized image) with a threshold value and performs binarization processing.
- the image I104 is an image (binarized image) in which a plurality of pixels constituting the image I103 (edge amount visualized image) are binarized.
- White pixels are pixels whose edge amount exceeds the threshold value
- black pixels are pixels whose edge amount is equal to or less than the threshold value.
- An image (a pseudo image of the moving object image) formed by the pixels constituting the moving object image included in the monitoring image is configured by the white pixels (pixels whose edge amount exceeds the threshold value).
- the correction amount calculation process and the monitoring image correction process described in FIG. 32 may be performed instead of the binarized image generation process.
- the fourth aspect of the present embodiment is the above-described image processing (edge amount calculation processing, binarization processing, correction amount calculation) for the maximum value visualized image (image I100 shown in FIG. 39) instead of the difference image. Processing, monitoring image correction processing).
- FIG. 41 is an image diagram showing correlation value visualized images arranged in time series, as in FIG.
- the monitoring target is a processing apparatus related to gas.
- the image I110 is an infrared image of the processing apparatus related to gas.
- the images I111 to I114 are four correlation value visualized images selected from the correlation value visualized images generated by using the moving image of the infrared image of the gas processing apparatus.
- the image I112 is an image after 2 seconds from the image I111
- the image I113 is an image after 2 seconds from the image I112
- the image I114 is an image after 2 seconds from the image I113.
- FIG. 42 is an image diagram showing various images related to the differential image generation process, as in FIG.
- the image I120 is a maximum value visualized image
- the image I121 is a minimum value visualized image
- the image I112 is a difference image.
- the edge noise of the image of the processing apparatus related to gas is weak.
- the arithmetic processing unit 9 processes predetermined information, and the image generation unit 8 visualizes the processed information.
- the display control unit 12 causes the display unit 13 to display the image.
- the present invention is not limited to this configuration, and includes the arithmetic processing unit 9.
- the configuration may not include the image generation unit 8, the display control unit 12, and the display unit 13, or the arithmetic processing unit 9 and the image generation unit 8 may be provided.
- the structure which does not include the display control part 12 and the display part 13 may be sufficient, and although the arithmetic processing part 9, the image generation part 8, and the display control part 12 are provided, the structure which does not include the display part 13 may be sufficient.
- the gas detection image processing apparatus forms an infrared image to be monitored with respect to an identification value for identifying a pixel constituting a gas image and a pixel constituting a non-gas image.
- a calculation unit that calculates the identification value corresponding to each of a plurality of pixels; and a monitor image that is generated using the infrared image based on the identification value,
- An identification unit that identifies pixels constituting the gas image. This corresponds to the first to fourth aspects of the embodiment.
- the 1st aspect of embodiment is comprised by step S100, S101, and S102, as shown in FIG.
- step S100 monitoring image generation processing
- step S101 gas concentration thickness product calculation processing
- [monitoring image generation processing] and [gas it is not necessary to read the “concentration / thickness product calculation process”, but only read the “discrimination process between the pixels constituting the gas image and the pixels constituting the moving object other than the gas image”. It is possible to understand the embodiments.
- the pixels constituting the gas image included in the monitoring image, and the non-gas image included in the monitoring image (for example, an image of a moving body other than the gas image). ) Can be identified, so that it is possible to identify a gas image and a non-gas image in the monitoring image.
- an image data input unit to which image data indicating a plurality of infrared images obtained by photographing the monitoring target at a plurality of times is input, and a plurality of images input from the image data input unit Generating time-series pixel data in which the pixel data of the pixels at the same position are arranged in time series in the infrared image, and obtaining the time-series pixel data of each of the plurality of pixels constituting the infrared image.
- This configuration corresponds to the first aspect of the embodiment.
- This configuration uses the gas concentration thickness product as the identification value.
- the determination unit is realized by the image generation unit 8 and the arithmetic processing unit 9 illustrated in FIG. 1A. Prepare a filter that transmits the wavelength range that is absorbed by the gas to be detected and a filter that does not transmit the wavelength range, and switch these filters appropriately to measure the background temperature with and without gas. There are known techniques. On the other hand, in this configuration, the background temperature with gas and the background temperature without gas are obtained by utilizing the phenomenon that the leaked gas fluctuates. According to this configuration, two types of filters and a mechanism for switching between them are not required.
- the identification value is a correlation value that correlates with the gas concentration thickness product.
- the correlation value is calculated using, for example, a first temperature difference generated by gas leaked from the monitoring target and a second temperature difference based on the temperature of the gas.
- the first temperature difference is a temperature difference between a background temperature with gas and a background temperature without gas
- the second temperature difference is a temperature difference between the temperature of the gas and the background temperature without gas
- the concentration thickness product may not be obtained from the relationship of gas temperature. This is inconvenient because it cannot be performed, for example, to perform a filtering process to see a spatial change on an image for determining whether the gas is a gas.
- the correlation value is defined by the above (formula)
- the background temperature with gas, the background temperature without gas, and the gas temperature are determined, the correlation value is determined. Therefore, in the correlation value visualized image, which is an image in which the correlation value is visualized, all the correlation values corresponding to each of the plurality of pixels constituting this image are determined. Therefore, all of the plurality of pixels constituting this image are used. There is no inconvenience when image processing is performed.
- the identification unit compares an absolute value of the identification value corresponding to each of the plurality of pixels constituting the infrared image with a predetermined threshold value, and exceeds the threshold value.
- a pixel corresponding to the absolute value of the identification value is specified as a pixel constituting the non-gas image.
- the pixel can be regarded not as a pixel constituting a gas image but as a pixel constituting a non-gas image.
- This configuration uses a threshold value to determine whether or not an identification value corresponding to a certain pixel is an abnormal value.
- the image processing device further includes an image generation unit that generates a visualized image that is an image obtained by visualizing the identification value calculated by the calculation unit, and the identification unit includes a plurality of pixels that form the visualized image.
- a spatial change amount (for example, an edge amount) of the identification value is calculated, and pixels constituting the non-gas image are specified based on the calculated spatial change amount.
- the amount of spatial change is a value obtained by spatially differentiating each of the identification values of the plurality of pixels constituting the visualized image.
- the identification unit uses a value obtained by spatially differentiating the identification value as a differential value, calculates the differential value of each of the plurality of pixels constituting the visualized image, and sets the calculated differential value to a predetermined threshold. A pixel corresponding to the differential value exceeding the threshold is specified as a pixel constituting the non-gas image.
- the identification unit corrects the identification value of the pixels constituting the non-gas image included in the monitoring image specified by the identification unit.
- This configuration corresponds to the second aspect of the embodiment.
- the monitor image after correction can suppress the brightness of the non-gas image compared to the brightness of the gas image.
- the image further includes an image generation unit that generates a plurality of the visualized images arranged in time series, using the image obtained by visualizing the identification value calculated by the calculation unit as a visualized image.
- a plurality of pixels constituting the visualized image by generating time-series identification value data in which the identification values corresponding to pixels at the same spatial position are arranged in time series in the plurality of visualized images arranged in series
- the pixels constituting the non-gas image corresponds to the third aspect of the embodiment.
- the identification unit is configured based on a temporal change amount of the identification value having the same value with respect to the time-series identification value data (for example, based on a histogram indicating the appearance frequency of the identification value having the same value). And specifying the pixels constituting the non-gas image.
- This configuration corresponds to the third aspect of the embodiment.
- the gas detection image processing method configures an infrared image to be monitored with respect to an identification value for identifying a pixel constituting a gas image and a pixel constituting a non-gas image.
- an identification value for identifying a pixel constituting a gas image and a pixel constituting a non-gas image.
- the gas detection image processing program configures an infrared image to be monitored with respect to an identification value for identifying a pixel constituting a gas image and a pixel constituting a non-gas image.
- an identification step of calculating the identification value corresponding to each of a plurality of pixels, and in a monitoring image generated using the infrared image based on the identification value the pixels constituting the gas image and the non-
- An identification step for identifying pixels constituting a gas image is executed by a computer.
- a computer-readable recording medium records the gas detection image processing program.
- An image processing method for gas detection according to a second aspect of the embodiment, an image processing program for gas detection according to the third aspect of the embodiment, and a computer-readable recording medium according to the fourth aspect of the embodiment An image processing apparatus for gas detection according to a first aspect of the embodiment is defined from the viewpoint of a method, a program, and a computer-readable recording medium, and the image processing apparatus for gas detection according to the first aspect of the embodiment It has the same effect.
- the gas detection system which concerns on 5th aspect of embodiment is an infrared camera which image
- the monitoring target area is, for example, an area where a gas leak monitoring target exists.
- the gas detection system which concerns on 5th aspect of embodiment is a system which calculates the identification value for identifying the pixel which comprises a gas image, and the pixel which comprises a non-gas image.
- the image processing apparatus for gas detection the image processing method for gas detection, the image processing program for gas detection, and the computer-readable recording medium which recorded the image processing program for gas detection can be provided. .
Abstract
Description
監視画像の生成方法として、様々な方法があるが、ここでは、監視画像の生成方法の一例を説明する。監視画像は、監視対象及び背景の赤外画像を利用して生成される。
赤外画像は、複数の画素が二次元に配列されて構成される。監視対象を含む背景は、複数の画素のそれぞれに対応する複数の領域に仮想的に分割されている。各画素の画素データは、対応する領域の背景温度を示している。ある領域に位置するガスの濃度厚み積を算出するためには、その領域にガスが有る場合のその領域の背景温度(ガス有り背景温度)、及び、その領域にガスが無い場合のその領域の背景温度(ガス無し背景温度)が必要となる。
図1Aを参照して、赤外線カメラ2で撮影された赤外画像の動画データD1が、ガス検知用画像処理装置3に送られる。図22は、赤外画像の動画データD1から選択された三つの赤外画像を示す画像図である。画像I30は、時刻T30に撮影された試験場所の赤外画像である。画像I31は、時刻T30から1秒後の時刻T31に撮影された試験場所の赤外画像である。画像I32は、時刻T30から2秒後に撮影された試験場所の赤外画像である。いずれの時刻でも、地点SP5でガスが噴出している。画像I30には写っていないが、画像I31の上部において、左から中央に向けて、動体(走行する電車)の像が写っている。画像I31が撮影された時刻T31から1秒経過後の時刻T32に撮影された画像I32の上部において、左から右に向けて、動体の像が写っている。
相関値=(ガス有り背景温度とガス無し背景温度との温度差)/(ガスの温度とガス無し背景温度との温度差)・・・(式)
実施形態の第1の局面に係るガス検知用画像処理装置は、ガス像を構成する画素と非ガス像を構成する画素とを識別するための識別値について、監視対象の赤外画像を構成する複数の画素のそれぞれに対応する前記識別値を算出する算出部と、前記識別値を基にして、前記赤外画像を用いて生成される監視画像において、前記ガス像を構成する画素と前記非ガス像を構成する画素とを識別する識別部と、を備える。これは、実施形態の第1態様~第4態様に対応する。なお、実施形態の第1態様は、図3に示すように、ステップS100、S101及びS102により構成される。ステップS100(監視画像の生成処理)及びステップS101(ガスの濃度厚み積の算出処理)については、公知の方法を用いることもできるので、明細書中の[監視画像の生成処理]及び[ガスの濃度厚み積の算出処理]の箇所を読まずに、[ガス像を構成する画素とガス像以外の動体の像を構成する画素との識別処理]の箇所を読むだけでも、実施形態の第1態様を理解することが可能である。
相関値=(ガス有り背景温度とガス無し背景温度との温度差)/(ガスの温度とガス無し背景温度との温度差)・・・(式)
Claims (18)
- ガス像を構成する画素と非ガス像を構成する画素とを識別するための識別値について、監視対象の赤外画像を構成する複数の画素のそれぞれに対応する前記識別値を算出する算出部と、
前記識別値を基にして、前記赤外画像を用いて生成される監視画像において、前記ガス像を構成する画素と前記非ガス像を構成する画素とを識別する識別部と、を備えるガス検知用画像処理装置。 - 前記識別値は、ガスの濃度厚み積である請求項1に記載のガス検知用画像処理装置。
- 前記監視対象が複数の時刻で撮影されることにより得られた、複数の前記赤外画像を示す画像データが入力される画像データ入力部と、
前記画像データ入力部から入力された複数の前記赤外画像において、同じ位置にある前記画素の画素データを時系列に並べた時系列画素データを生成し、前記赤外画像を構成する前記複数の画素のそれぞれの前記時系列画素データを基にして、前記複数の画素のそれぞれに対応する、ガスが有る場合の背景温度を示すガス有り背景温度及びガスが無い場合の背景温度を示すガス無し背景温度を決定する決定部と、をさらに備え、
前記算出部は、前記決定部によって決定された前記ガス有り背景温度及び前記ガス無し背景温度を利用して、前記複数の画素のそれぞれに対応する前記濃度厚み積を算出する請求項2に記載のガス検知用画像処理装置。 - 前記識別値は、ガスの濃度厚み積と相関する相関値である請求項1に記載のガス検知用画像処理装置。
- 前記相関値は、前記監視対象から漏れたガスにより発生する第1の温度差と、前記ガスの温度を基準にした第2の温度差とを用いて算出される請求項4に記載のガス検知用画像処理装置。
- 前記第1の温度差は、ガス有り背景温度とガス無し背景温度との温度差であり、前記第2の温度差は、前記ガスの温度と前記ガス無し背景温度との温度差であり、前記相関値は、下記式で示される請求項5に記載のガス検知用画像処理装置。
相関値=(ガス有り背景温度とガス無し背景温度との温度差)/(ガスの温度とガス無し背景温度との温度差)・・・(式) - 前記識別部は、前記赤外画像を構成する前記複数の画素のそれぞれに対応する前記識別値の絶対値を、予め定められたしきい値と比較し、前記しきい値を超えている前記識別値の絶対値に対応する画素を、前記非ガス像を構成する画素として特定する請求項1~6のいずれか一項に記載のガス検知用画像処理装置。
- 前記算出部によって算出された前記識別値を可視化した画像である可視化画像を生成する画像生成部をさらに備え、
前記識別部は、前記可視化画像を構成する複数の画素のそれぞれについて、前記識別値の空間的変化量を算出し、算出した前記空間的変化量を基にして、前記非ガス像を構成する画素を特定する請求項1~6のいずれか一項に記載のガス検知用画像処理装置。 - 前記空間的変化量は、前記可視化画像を構成する前記複数の画素のそれぞれの前記識別値を空間微分した値であり、
前記識別部は、前記識別値を空間微分した値を微分値とし、前記可視化画像を構成する前記複数の画素のそれぞれの前記微分値を算出し、算出した前記微分値を予め定められたしきい値と比較し、前記しきい値を超えている前記微分値に対応する画素を、前記非ガス像を構成する画素として特定する請求項8に記載のガス検知用画像処理装置。 - 前記識別部は、前記識別部によって特定された、前記監視画像に含まれる前記非ガス像を構成する画素の前記識別値を小さくする補正をする請求項9に記載のガス検知用画像処理装置。
- 前記算出部によって算出された前記識別値を可視化した画像を可視化画像とし、時系列に並べられた複数の前記可視化画像を生成する画像生成部をさらに備え、
前記算出部は、時系列に並べられた複数の前記可視化画像において、空間的に同じ位置にある画素に対応する前記識別値を時系列に並べた時系列識別値データを生成し、
前記識別部は、前記可視化画像を構成する複数の画素のそれぞれについて、前記時系列識別値データの中に異常な値の前記識別値を含むか否かを判定し、前記異常な値の前記識別値を含む前記時系列識別値データに対応する画素を、前記非ガス像を構成する画素として特定する請求項1~6のいずれか一項に記載のガス検知用画像処理装置。 - 前記識別部は、前記時系列識別値データについて、同じ値の前記識別値の時間的変化量を基にして、前記非ガス像を構成する画素を特定する請求項11に記載のガス検知用画像処理装置。
- ガス像を構成する画素と非ガス像を構成する画素とを識別するための識別値について、監視対象の赤外画像を構成する複数の画素のそれぞれに対応する前記識別値を算出する算出ステップと、
前記識別値を基にして、前記赤外画像を用いて生成される監視画像において、前記ガス像を構成する画素と前記非ガス像を構成する画素とを識別する識別ステップと、を備えるガス検知用画像処理方法。 - ガス像を構成する画素と非ガス像を構成する画素とを識別するための識別値について、監視対象の赤外画像を構成する複数の画素のそれぞれに対応する前記識別値を算出する算出ステップと、
前記識別値を基にして、前記赤外画像を用いて生成される監視画像において、前記ガス像を構成する画素と前記非ガス像を構成する画素とを識別する識別ステップと、をコンピュータに実行させるガス検知用画像処理プログラム。 - 請求項14に記載されたガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体。
- 監視対象領域を撮影する赤外線カメラと、
ガス像を構成する画素と非ガス像を構成する画素とを識別するための識別値について、前記赤外線カメラによって撮影された前記監視対象領域の画像を構成する複数の画素のそれぞれに対応する前記識別値を算出する算出部と、を備えるガス検知システム。 - 前記識別値は、ガスの濃度厚み積である請求項16に記載のガス検知システム。
- 前記識別値は、ガスの濃度厚み積と相関する相関値である請求項16に記載のガス検知システム。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017556045A JP6319527B2 (ja) | 2015-12-15 | 2016-12-12 | ガス検知用画像処理装置、ガス検知用画像処理方法、ガス検知用画像処理プログラム、ガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、及び、ガス検知システム |
US16/062,622 US10852213B2 (en) | 2015-12-15 | 2016-12-12 | Image processing device for gas detection, image processing method for gas detection, image processing program for gas detection, computer-readable recording medium having image processing program for gas detection recorded thereon, and gas detection system |
EP16875594.0A EP3392635B1 (en) | 2015-12-15 | 2016-12-12 | Image processing device for gas detection, image processing method for gas detection, image processing program for gas detection, computer-readable recording medium having image processing program for gas detection recorded thereon, and gas detection system |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015244193 | 2015-12-15 | ||
JP2015-244193 | 2015-12-15 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017104617A1 true WO2017104617A1 (ja) | 2017-06-22 |
Family
ID=59056663
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2016/086929 WO2017104617A1 (ja) | 2015-12-15 | 2016-12-12 | ガス検知用画像処理装置、ガス検知用画像処理方法、ガス検知用画像処理プログラム、ガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、及び、ガス検知システム |
Country Status (4)
Country | Link |
---|---|
US (1) | US10852213B2 (ja) |
EP (1) | EP3392635B1 (ja) |
JP (1) | JP6319527B2 (ja) |
WO (1) | WO2017104617A1 (ja) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017213075A1 (ja) * | 2016-06-07 | 2017-12-14 | コニカミノルタ株式会社 | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム |
CN108071422A (zh) * | 2018-02-23 | 2018-05-25 | 中国矿业大学(北京) | 基于图像监测设备的矿井爆炸监控系统 |
CN108071423A (zh) * | 2018-02-23 | 2018-05-25 | 中国矿业大学(北京) | 基于红外图像监测设备的矿井爆炸监控系统 |
CN108131166A (zh) * | 2018-02-23 | 2018-06-08 | 中国矿业大学(北京) | 基于图像的矿井爆炸监测报警系统 |
CN108590763A (zh) * | 2018-02-23 | 2018-09-28 | 中国矿业大学(北京) | 基于红外图像的矿井爆炸监测报警系统 |
WO2019138873A1 (ja) * | 2018-01-09 | 2019-07-18 | コニカミノルタ株式会社 | ガス検知用画像処理装置およびガス検知用画像処理方法 |
JP2019184485A (ja) * | 2018-04-13 | 2019-10-24 | コニカミノルタ株式会社 | ガス検知用画像処理装置、ガス検知用画像処理方法、及び、ガス検知用画像処理プログラム |
US10997734B2 (en) * | 2016-12-27 | 2021-05-04 | Konica Minolta, Inc. | Gas detection-use image processing device, gas detection-use image processing method, and gas detection-use image processing program |
WO2022264604A1 (ja) | 2021-06-16 | 2022-12-22 | コニカミノルタ株式会社 | ガス濃度特徴量推定装置、ガス濃度特徴量推定方法、プログラム、およびガス濃度特徴量推論モデル生成装置 |
WO2023105856A1 (ja) * | 2021-12-10 | 2023-06-15 | コニカミノルタ株式会社 | ガス濃度測定装置、ガス濃度測定方法、およびプログラム |
JP7438850B2 (ja) | 2020-05-29 | 2024-02-27 | キヤノンメディカルシステムズ株式会社 | 医用画像診断装置及び医用画像処理装置 |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102017010151A1 (de) * | 2017-11-02 | 2019-05-02 | Dräger Safety AG & Co. KGaA | lnfrarot-optische Gasmessvorrichtung |
US11466554B2 (en) | 2018-03-20 | 2022-10-11 | QRI Group, LLC | Data-driven methods and systems for improving oil and gas drilling and completion processes |
US11506052B1 (en) | 2018-06-26 | 2022-11-22 | QRI Group, LLC | Framework and interface for assessing reservoir management competency |
US11108995B2 (en) * | 2018-09-11 | 2021-08-31 | Draeger Medical Systems, Inc. | System and method for gas detection |
US11529979B2 (en) | 2020-03-10 | 2022-12-20 | Progress Rail Services Corporation | Systems and methods for adjusting train operation |
DE102021100405A1 (de) | 2021-01-12 | 2022-07-14 | Inficon Gmbh | Verfahren zur Detektion von aus einem Prüfling austretendem Prüfgas mit einem optischen Sensor |
KR102539500B1 (ko) * | 2022-10-21 | 2023-06-05 | (주)에스에프테크놀로지 | 하나의 장치에서 유해가스 탐지와 방사선 탐지로 이루어진 원샷 하이브리드형 화생방 통합 원격 탐지 장치 및 방법 |
CN115841434B (zh) * | 2023-02-21 | 2023-04-21 | 深圳市特安电子有限公司 | 一种用于气体浓度分析的红外图像增强方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5215385B2 (ja) * | 1973-08-15 | 1977-04-28 | ||
JPS6232336A (ja) * | 1985-08-05 | 1987-02-12 | Toshiba Corp | Itvカメラによる液体漏洩検出方法 |
JPH07210784A (ja) * | 1994-01-20 | 1995-08-11 | Hitachi Ltd | 監視装置 |
JP2012058093A (ja) * | 2010-09-09 | 2012-03-22 | Mitsubishi Electric Building Techno Service Co Ltd | ガス漏れ検出装置 |
US20140002639A1 (en) * | 2011-03-25 | 2014-01-02 | Joseph M. Cheben | Autonomous Detection of Chemical Plumes |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5215386A (en) * | 1975-07-25 | 1977-02-04 | Babcock Hitachi Kk | Rotor flaw detection method by acoustic emission signal |
JPS5215385A (en) | 1975-07-25 | 1977-02-04 | Hitachi Ltd | Image recognizer |
EP0544962B1 (fr) * | 1991-12-04 | 1997-03-05 | Bertin & Cie | Procédé et dispositif de détection optique à distance d'un gaz présent dans une zone d'espace observée |
JPH06288858A (ja) * | 1993-03-31 | 1994-10-18 | Osaka Gas Co Ltd | ガスの可視化装置 |
SE9904836L (sv) | 1999-12-28 | 2001-06-29 | Jonas Sandsten | Kvantitativ avbildning av gasemissioner utnyttjande optisk teknik |
JP3776681B2 (ja) * | 2000-06-06 | 2006-05-17 | ホーチキ株式会社 | ガス監視装置 |
US6822742B1 (en) | 2003-12-19 | 2004-11-23 | Eastman Kodak Company | System and method for remote quantitative detection of fluid leaks from a natural gas or oil pipeline |
JP5573340B2 (ja) | 2010-05-07 | 2014-08-20 | 日本電気株式会社 | ガス検知装置およびガス検知方法 |
JP2012220313A (ja) * | 2011-04-07 | 2012-11-12 | Mitsubishi Electric Corp | 気体検出装置 |
CN103217397A (zh) * | 2013-01-23 | 2013-07-24 | 山西省电力公司晋城供电分公司 | 基于红外图像处理的sf6气体检测方法 |
US9488629B2 (en) | 2014-04-10 | 2016-11-08 | General Electric Company | Method for detecting coolant leaks in generators |
WO2015172056A1 (en) * | 2014-05-09 | 2015-11-12 | Kairos Aerospace Inc. | Systems and methods for detecting gas leaks |
-
2016
- 2016-12-12 US US16/062,622 patent/US10852213B2/en active Active
- 2016-12-12 JP JP2017556045A patent/JP6319527B2/ja active Active
- 2016-12-12 WO PCT/JP2016/086929 patent/WO2017104617A1/ja active Application Filing
- 2016-12-12 EP EP16875594.0A patent/EP3392635B1/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5215385B2 (ja) * | 1973-08-15 | 1977-04-28 | ||
JPS6232336A (ja) * | 1985-08-05 | 1987-02-12 | Toshiba Corp | Itvカメラによる液体漏洩検出方法 |
JPH07210784A (ja) * | 1994-01-20 | 1995-08-11 | Hitachi Ltd | 監視装置 |
JP2012058093A (ja) * | 2010-09-09 | 2012-03-22 | Mitsubishi Electric Building Techno Service Co Ltd | ガス漏れ検出装置 |
US20140002639A1 (en) * | 2011-03-25 | 2014-01-02 | Joseph M. Cheben | Autonomous Detection of Chemical Plumes |
Non-Patent Citations (1)
Title |
---|
See also references of EP3392635A4 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10908079B2 (en) | 2016-06-07 | 2021-02-02 | Konica Minolta, Inc. | Image-processing device for gas detection, image-processing method for gas detection, and image-processing program for gas detection |
WO2017213075A1 (ja) * | 2016-06-07 | 2017-12-14 | コニカミノルタ株式会社 | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム |
US10997734B2 (en) * | 2016-12-27 | 2021-05-04 | Konica Minolta, Inc. | Gas detection-use image processing device, gas detection-use image processing method, and gas detection-use image processing program |
WO2019138873A1 (ja) * | 2018-01-09 | 2019-07-18 | コニカミノルタ株式会社 | ガス検知用画像処理装置およびガス検知用画像処理方法 |
US11519852B2 (en) | 2018-01-09 | 2022-12-06 | Konica Minolta, Inc. | Gas detection-use image processing device, and gas detection-use image processing method |
CN108590763A (zh) * | 2018-02-23 | 2018-09-28 | 中国矿业大学(北京) | 基于红外图像的矿井爆炸监测报警系统 |
CN108131166A (zh) * | 2018-02-23 | 2018-06-08 | 中国矿业大学(北京) | 基于图像的矿井爆炸监测报警系统 |
CN108071423A (zh) * | 2018-02-23 | 2018-05-25 | 中国矿业大学(北京) | 基于红外图像监测设备的矿井爆炸监控系统 |
CN108071422A (zh) * | 2018-02-23 | 2018-05-25 | 中国矿业大学(北京) | 基于图像监测设备的矿井爆炸监控系统 |
CN108131166B (zh) * | 2018-02-23 | 2023-04-14 | 中国矿业大学(北京) | 基于图像的矿井爆炸监测报警系统 |
CN108071422B (zh) * | 2018-02-23 | 2023-04-14 | 中国矿业大学(北京) | 基于图像监测设备的矿井爆炸监控系统 |
CN108590763B (zh) * | 2018-02-23 | 2023-04-14 | 中国矿业大学(北京) | 基于红外图像的矿井爆炸监测报警系统 |
CN108071423B (zh) * | 2018-02-23 | 2023-04-14 | 中国矿业大学(北京) | 基于红外图像监测设备的矿井爆炸监控系统 |
JP2019184485A (ja) * | 2018-04-13 | 2019-10-24 | コニカミノルタ株式会社 | ガス検知用画像処理装置、ガス検知用画像処理方法、及び、ガス検知用画像処理プログラム |
JP7056342B2 (ja) | 2018-04-13 | 2022-04-19 | コニカミノルタ株式会社 | ガス検知用画像処理装置、ガス検知用画像処理方法、及び、ガス検知用画像処理プログラム |
JP7438850B2 (ja) | 2020-05-29 | 2024-02-27 | キヤノンメディカルシステムズ株式会社 | 医用画像診断装置及び医用画像処理装置 |
WO2022264604A1 (ja) | 2021-06-16 | 2022-12-22 | コニカミノルタ株式会社 | ガス濃度特徴量推定装置、ガス濃度特徴量推定方法、プログラム、およびガス濃度特徴量推論モデル生成装置 |
WO2023105856A1 (ja) * | 2021-12-10 | 2023-06-15 | コニカミノルタ株式会社 | ガス濃度測定装置、ガス濃度測定方法、およびプログラム |
Also Published As
Publication number | Publication date |
---|---|
EP3392635B1 (en) | 2022-06-08 |
JPWO2017104617A1 (ja) | 2018-03-01 |
US20190003919A1 (en) | 2019-01-03 |
US10852213B2 (en) | 2020-12-01 |
EP3392635A1 (en) | 2018-10-24 |
EP3392635A4 (en) | 2019-05-08 |
JP6319527B2 (ja) | 2018-05-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6319527B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法、ガス検知用画像処理プログラム、ガス検知用画像処理プログラムを記録したコンピュータ読み取り可能な記録媒体、及び、ガス検知システム | |
JP6344533B2 (ja) | ガス濃度厚み積測定装置、ガス濃度厚み積測定方法、ガス濃度厚み積測定プログラム、及び、ガス濃度厚み積測定プログラムを記録したコンピュータ読み取り可能な記録媒体 | |
US11092507B2 (en) | Method of indicating gas movement in a scene | |
JP6245418B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム | |
JP6620878B2 (ja) | ガス漏れ位置推定装置、ガス漏れ位置推定方法及びガス漏れ位置推定プログラム | |
JP6468439B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム | |
JP6579290B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法、及び、ガス検知用画像処理プログラム | |
JP6508439B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム | |
US11393096B2 (en) | Gas-detection image processing device, gas-detection image processing method, and gas-detection image processing program | |
JP2020071698A (ja) | 火災検知装置、火災検知方法及び火災監視システム | |
JP7047638B2 (ja) | ガス可視化用画像処理装置、ガス可視化用画像処理方法、ガス可視化用画像処理プログラム、及び、ガス検知システム | |
US20200258267A1 (en) | Image processing device for gas detection, image processing method for gas detection, and image processing program for gas detection | |
JP7156291B2 (ja) | ガス検査レポート作成支援装置、ガス検査レポート作成支援方法、及び、ガス検査レポート作成支援プログラム | |
US20220034742A1 (en) | Gas flow rate estimation device, gas flow rate estimation method, and gas flow rate estimation program | |
JP6504325B2 (ja) | ガス検知用画像処理装置、ガス検知用画像処理方法及びガス検知用画像処理プログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16875594 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017556045 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2016875594 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2016875594 Country of ref document: EP Effective date: 20180716 |