WO2015166265A1

WO2015166265A1 - Self-correcting gas camera

Info

Publication number: WO2015166265A1
Application number: PCT/GB2015/051279
Authority: WO
Inventors: Roger HUTTON; Paul Basham
Original assignee: Crowcon Detection Instruments Limited
Priority date: 2014-05-01
Filing date: 2015-05-01
Publication date: 2015-11-05
Also published as: GB2526066A; GB201407677D0

Abstract

An imaging device (100) for use in gas detection comprising a thermographic camera (102) and a sensor (104) for measuring temperature, and a processor (106). The device is configured to generate a first thermal image of radiation of a first location using the camera (S300), measure an ambient temperature of the location using the sensor (S302), calculate a first value based on the temperature of a first portion of the first image and the measured ambient temperature (S306), compare the first value of the first portion of the first image to a predetermined value indicative of a detection limit of the camera (S308, S310), and assign a level of confidence to a first area of the location corresponding to the first portion of the first image based on the comparison (S310, S314).

Description

Self-Correcting Gas Camera

Technical field of invention

The invention relates to improvements in imaging devices for use in gas detection, in particular passive optical imaging devices.

Background to the invention

The detection and monitoring of gas leaks and the build-up of gas is important in many industries, particularly in those involved in the processing of oil, gas and chemicals. In such industries, gas leaks can present a range of hazards depending on the nature of the gas, including harm to persons and the environment due to toxicity, and risk of explosion.

In response to this need, a number of different technologies have been developed to detect the presence of gases. One such technology uses passive optical imaging to detect potential gas leaks in an area corresponding to the field of view a thermographic camera. It is known for such a camera to image the thermal radiation emitted by background objects and structures in the field of view. Such cameras detect thermal radiation emitted by the background, and the subsequent absorption of the thermal radiation by gas in the field of view of the camera thus allowing for the presence of a gas to be detected. In such systems a gas leak may be imaged in real time. As different gases have unique optical absorption spectra, it is known to fit optical filters to the camera to allow different wavelength bands of radiation to be imaged in turn. Depending on the relative amount of radiation of each wavelength absorbed in the region of an image corresponding to a gas leak, the chemical composition of the gas can be determined.

However for the above technique to work effectively, passive optical thermal imaging devices require there to be a significant temperature difference between the background and gas to be detected. The exact value of this difference is related to a detection limit of the camera. For example, in order to detect a 10% lower explosion limit concentration (LEL.m) of methane gas, a temperature difference between the background and the gas of around 2K is typically required, depending on the characteristics of the particular camera being used. When the temperature difference between the background and environment is lower than the detection limit of the camera (i.e. when ΔΤ, ΔΤ being the difference between the background and environment temperature, is small), environmental changes, such as the presence of fog, may lead to false positive detections.

Summary of invention

In order to mitigate at least some of the above problems, the present invention provides an imaging device for use in gas detection comprising a thermographic camera and a sensor for measuring the ambient temperature. The device is configured to generate a first thermal image of radiation of a first location using the camera, measure an ambient temperature of the location using the sensor, calculate a first value based on the temperature of a first portion of the first image and the measured ambient temperature, compare the first value of the first portion of the first image to a predetermined value indicative of a detection limit of the camera, and assign a level of confidence to an area of the location corresponding to the first portion of the first image based on the comparison.

Therefore, the present invention identifies situations where the difference between the background and the atmosphere is low which may result in false positive readings and ultimately allow for remedial action to be taken. As well as providing a measure of confidence to the various regions/locations of the image the invention allows for the easy identification of such regions and enable an end user to take remedial action. In industrial environments such as gas plants, refineries etc., the ability to take such remedial action helps ensure the safety of persons within the environment.

In a preferred embodiment device is further configured to detect gas using known gas detection analysis based on the level of confidence on the first area. For example in one embodiment, if the level of confidence for a certain area or gas cloud falls below a detection threshold value, the area is excluded from gas detection analysis. In this embodiment the device is further configured to start gas detection analysis if the level of confidence of the area subsequently exceeds said detection threshold.

Preferably the first value is the difference in temperature between the ambient atmosphere and the background (which may constitute permanent structures or landscape), ΔΤ, although the skilled person will appreciate other values based on the temperature of the ambient air and of the background could also be used for the same purpose. Preferably the imaging device is configured to be used to survey one or more locations to identify regions or structures corresponding to a given level of confidence. For example, structures with a surface of a certain reflectivity may have a small value for ΔΤ, making detection of target gas leaks difficult using thermographic cameras. The imaging device would then indicate such surfaces to a user. The user would then be able to take appropriate action, for example painting the surface identified so as to adjust its reflectivity.

In a preferred embodiment the device further comprises a display, configured to display a visual image corresponding to the radiation imaged by the camera. Such a display allows a user to visually identify gas leaks quickly, in addition to gas detection analysis performed by the device itself, thus providing an additional level of safety.

In a preferred embodiment the device is configured to display a visual indication when an area has a level of confidence below a certain threshold value, and indicate the part of the visual image corresponding to the area on the display. Advantageously this allows a user to easily identify when false negative gas detection results may be obtained, in which area this may occur and take any necessary further action. In a further embodiment the device is also configured to issue an audible indication. In another embodiment only an audible indicator is provided by the device.

In a preferred embodiment the device is further configured to take a sequence of images to allow the area to be monitored in real time. In one embodiment, the camera may take images at a rate of 24 frames per second, alternatively at 30 frames per second, alternatively at 50 frames per second.

In another embodiment, the device is further configured to monitor the position of a non- static area corresponding to a given level of confidence in a sequence of images, and predict the location of the area in subsequent images. In a further embodiment the device can be configured to cease gas detection analysis when the area enters the field of view of the camera and commence gas detection analysis once the area has left the field of view. For example, such an area might correspond to cloud of fog, which leads to a ΔΤ being too low for reliable gas detection in the field of view, resulting in the device excluding the area when performing gas detection analysis. The device may predict how long the cloud of fog will obscure the view of the camera, and may further indicate this time to the user. Once the cloud of fog has left the field of view of the camera, gas detection analysis may include the area. By informing the user how long gas detection analysis will be unavailable, the user can take appropriate action, for example limiting access to the area by personnel. Furthermore, as the user knows when the gas detection analysis will be able to start in the area in advance, the time for which access to the area is limited can be minimised, thus reducing the impact on work done by other personnel.

In a preferred embodiment, the device further comprises a plurality of optical filters, wherein each filter is configured to prevent different selections of wavelengths being imaged by the camera, thus allowing the device to distinguish the chemical composition of gas being detected in a known way. In some embodiments, the filters are housed within the camera body itself. In one embodiment a different confidence level is also based on potential gas to be detected. For example, for the same ΔΤ, the level of confidence may be higher for gas A than for gas B. In a further embodiment, when the level of confidence for gas A falls below a detection threshold but the level of confidence for gas B does not, gas detection analysis for gas A is ceased, but is continued for gas B. In a further embodiment, the device chooses one or more filters for use based on gases for which the level of confidence is above a certain detection threshold.

In another embodiment the device is further configured to provide a record of one or more of gas detection analysis data, level of confidence data, ambient temperature data, and background temperature data. Advantageously, this allows a user to analyse gas leak events retrospectively to assist in finding the location of leak sites. It also allows the user to retrospectively analyse areas susceptible to low confidence levels, for example during adverse weather conditions. In some embodiments, the device further comprises memory, onto which the above data may be recorded. Such memory may be removable flash memory.

In a further embodiment the device may further comprise wireless communication circuitry. Such circuitry beneficially allows one or more such devices to be remotely operated by an electronic device from a central location. This is particularly useful in situations when gas detection must be performed in hard to access locations. Furthermore wireless communication circuitry also allows the data discussed above to be downloaded to a centrally located data recording device. In another embodiment, the device comprises wired communication circuitry to allow communication over a wired connection instead providing the same benefits. In a further embodiment, the device comprises both wired and wireless communication circuitry.

In some embodiments a remote heat source is provided. Such a heat source is positioned in the field of view of the thermographic camera, and is configured to generate thermal radiation of a known temperature. Preferably the heat source is configured to generate radiation across a range of temperatures, which the user can select between. Advantageously the heat source can be used to artificially generate a value of ΔΤ for a particular location that is large enough to allow effective gas detection. Therefore the heat source may be used to ensure a particular location is assigned a high level of confidence.

In further embodiments, the heat source is used to calibrate the imaging device by comparing the temperature of the heat source radiation as measured by the thermographic camera to the known value of the temperature of the heat source radiation. Preferably the calibration is repeated for heat source radiation at a plurality of different temperatures.

Other aspects of the invention will become apparent from the appended claim set.

Brief description of the figures

Figure 1 is a schematic of an imaging device for use in gas detection in accordance with the present invention;

Figure 2 is a diagram illustrating the imaging device in operation; Figure 3 is a flow diagram outlining the method of determining the reliability of gas detection for an area imaged by the imaging device;

Figure 4 is a flow diagram outlining the method of operating the imaging device when it is determined that gas detection is unreliable; and Figure 5 is a flow diagram outlining the method of operating the imaging device when it is determined that gas detection is reliable. Description of an embodiment of the invention

Figure 1 shows a schematic of an imaging device for gas detection 100. The imaging device comprises: a thermographic camera 102 configured to image thermal radiation comprising optics 103; a temperature sensor 104 configured to measure ambient air temperature; and a processor 106 configured to perform gas detection analysis as described below. The imaging device further comprises a plurality of optical filters 108, communication circuitry 110, memory 112, and a display 114.

The temperature sensor 104 in a preferred embodiment is a thermocouple, though in further embodiments other temperature sensors may be used such as ptlOO devices or other IR thermal devices. In further embodiments the temperature sensor 104 is remote from the imaging device 100, in an example the temperature sensor is a remote thermocouple that wirelessly communicates with the imaging device.

The plurality of optical filters 108, held in an array, are provided so as to selective filter the radiation so as to enable the detection of certain gas types as is known in the art. In one embodiment the array is configured to be moved between a first position (not shown) and a second position (as shown in figure 2). When in the first position, light does not pass through the filters 108 and therefore the filters 108 do not affect the radiation imaged by the camera 102. When in the second position, the filters 108 are disposed so as to cover the optics 103 of the camera 102. The wavelength, and type, of filters is selected according to the types of gas to be detected. The selection of such filters is known in the art.

Alternatively the filters 108 do not move and are instead disposed such that different filters cover different parts of the sensor of the camera 102, thus different parts of the sensor array detect radiation having different bands of wavelengths. In this case the filters are preferably disposed between the optics 103 and the sensor array of the camera (not shown).

The imaging device 100 also comprises communication circuitry 110 configured to allow the transfer of data between the imaging device and an external/remote electronic device (not shown). The communication circuitry 110 is configured to allow both wired and wireless communication, preferably using known networking protocols such as HTTP or TCP. This beneficially allows the imaging device 100 to be remotely operated by the electronic device from a central location, for example a control room in a refinery. This is particularly useful in situations when gas detection must be performed in hard to access locations. Furthermore communication circuitry 110 also allows data to be sent from the imaging device to the electronic device. It is to be appreciated that other protocols may be used, and that communication circuitry 110 might only allow wired or wireless communication. Typically the camera is installed in industrial locations such as gas plants, refineries, oil rigs etc., where the camera is configured to view an extended area (several tens to hundreds of metres in size). In such environments it is known to have a central office, or control room, which monitor the safety of the industrial location and accordingly information from the camera in such embodiments is transmitted to the central location.

Memory 112 is included in the imaging device. Memory 112 may be present at the device itself, or at a remote location, the imaging device being configured to transmit data to the memory using communication circuitry 110. Memory 112 may take the form of nonremovable memory, such as a hard disc drive (HDD) or solid state drive (SSD), removable memory, such as secure digital (SD) cards or USB flash drives, or both. Advantageously memory 112 allows data generated by the imaging device 100 to be recorded. Such data may be used retrospectively to analyse gas leak events, the analysis being performed at either the imaging device 100 or at a remote location. The imaging device optionally comprises a display 114. The display is configured to display a visual image corresponding to the radiation imaged by the camera 102. Such a display allows a user to visually identify gas leaks quickly, in addition to gas detection analysis performed by the device itself, thus providing an additional level of safety. In some examples the display 114 is configured to communicate with communication circuitry 110 via a wired or wireless connection and may be positioned remotely from the camera. Alternatively the display 114 may be fixed to the imaging device 100. Alternatively both a fixed display and a remote display may be provided. For example, the imaging device may be configured to transmit data to one or more electronic devices at a remote location, such as a central control room. The control room would typically house a plurality of displays corresponding to a plurality of imaging devices, allowing a user to monitor multiple remote locations at once. The control room might also house the memory on which data is recorded.

Figure 2 shows the imaging device 100 in use, detecting a target gas leak from container 200. The camera 102 images thermal radiation emitted by background objects and structures 206. The processor calculates the blackbody temperature T_B 208 for each part of the image based on the wavelength of radiation detected.

Figure 2 shows a cloud of target gas 202 leaking from container 200. The target gas cloud 202 absorbs certain wavelengths of radiation emitted by the background objects 206 and emits characteristic radiation related to the temperature of the target gas TG 204. Processor 106 is configured perform gas detection analysis whereby the change in temperature of the radiation imaged in certain parts of the image when the target gas enters the field of view of the camera 102 is detected. Processor 106 then causes the display 114 to display an indication that a target gas 202 has been detected and an indication of the part of the image in which the target gas 202 has been detected. The processor may also cause a speaker (not shown) to issue an audible warning. The display of the indications above and the issuance of an audible warning may together form an alert system as known in the art.

For example, on detection of a target gas, a known alert system housed in a control room may be used to alert a user, or users, to the detection of the target gas.

The imaging device 100 is also able to identify the chemical composition of the target gas cloud 202. One or more of the optical filters 108 may be moved from the first position to the second position. The processor 106 is configured to analyse the relative amount of radiation of certain wavelengths being imaged in the presence of the different filters 108, compare the resulting spectrum to known gases and thus determine the composition of the target gas 202. It is noted that the identification of gas composition using different filters in this way is already known. The imaging device 100 is configured to determine a level of confidence on which to base gas detection analysis. Beneficially, the device 100 is configured to determine a level of confidence for individual pixels, or groups of pixels, (for example 2x2, 4x4 etc.) as well as for the entire image. Therefore in a given image the level of confidence associated with a detection may vary.

The ability to determine the level of confidence, in particular in situations where ΔΤ is small (near or below the detection limit of the camera) is particularly important in industrial situations where the field of view of the camera is large.

For example, the camera would typically be installed in industrial locations such as a gas processing plant/ refinery. Such refineries have a number of structures, including refrigeration/condensing structures, storage tanks as well as buildings. Across such a field of view, as imaged by the camera, there would typically be a variation of background temperatures due to the structures (e.g. a refrigeration structure), the emissivity of the structures (for example due to the material used in the structure, the type of paint and colour of the structure), as well as any natural temperature differences due to the blue sky background. In such situations the presence of a dangerous gas may occur at any location within the field of view. Accordingly due to the differences in the background temperatures the ability to detect the presence of a dangerous gas may vary across the field of view of the camera. Whilst it is possible to assign a confidence level regarding the detection of the gas to a whole image, this may result in an entire image (i.e. across the whole field of view of the camera) being deemed to have a high level of confidence where in fact due to the presence of, say a refrigeration tower, parts of the image should in fact be assigned a low level of confidence (due to the low ΔΤ). Similarly, an entire image may be deemed unreliable due to the low ΔΤ whereas a portion of the image may in fact be used to reliably detect the presence of gas (within said region of the image having a ΔΤ above the detection limit of the camera).

Accordingly, a beneficial aspect of the invention is the ability to compensate for the variations of background temperature across the whole image, in particular where the variations result in one or regions of low ΔΤ (i.e. regions which may be deemed unreliable). For a thermographic camera with a given detection limit, its ability to detect a target gas is related to the relative temperature of the target gas and radiation from the background. The present invention compares the ambient temperature 210 of a location to the background temperature T_B 208 in order to identify whether the temperatures of a target gas TG 204 and of the background T_B 208 are sufficiently different for effective gas detection for a given region. Beneficially is repeated across the entire image (i.e. for multiple regions) thereby enabling the determination as to which regions of the image allow for effective gas detection across the entire field of view of the camera.

The imaging device optionally comprises a remote heat source (not shown), which may be positioned in the field of view of the camera 102. The heat source is configured to generate thermal radiation across a range of known temperatures, which the user can select between. The heat source is used to calibrate the imaging device 100 by comparing the temperature of the heat source radiation as measured by the thermographic camera to the known value of the temperature of the heat source radiation. If the two values are different, the processor 106 corrects the output of the camera 102 such that the two temperatures are the same. Preferably the calibration is repeated for heat source radiation of different temperatures. Such calibration of the camera further helps ensure the accuracy of the camera over an extended period of time. In further embodiments, where the difference between the temperature of the heat source and the temperature as measured by the camera, is greater than a predetermined limit the processor 106 is further configured to emit a signal to indicate that the camera 102 is indeed of maintenance or repair. The signal, in an embodiment, is a warning message which is displayed on the display associated with the camera.

Figure 3 is a flowchart of the method used by imaging device 100 to determine the reliability of gas detection for each part of an image of thermal radiation imaged by the camera 102.

The following methodology is discussed with reference to a fixed single camera. The methodology may be extended to systems comprising multiple imaging devices 100, as well as image devices 100 which scans across an area, or portable imaging devices 100 which may be moved and placed in areas of interest. At step S300, the thermographic camera 102 generates a thermal image of the location corresponding to its field of view. The generation of the thermal image occurs in a known manner. When optical filters 108 are used, the camera 102 generates a different thermal image for each filter used.

At step S302, the imaging device 100 measures the ambient temperature TA 210 using temperature sensor 104. The ambient temperature is periodically updated, preferably the temperature is measured for each instance of a generation of a thermal image as per step S300.

At step S304, the imaging device 100 then uses processor 106 to determine the blackbody temperature T_B 208 of each part of the image corresponding to the wavelength of radiation emitted by the background objects and structures 206 in the field of view of the camera 102. In the present embodiment, each part of the image corresponds to a group of pixels (such as a 2x2, 4x4 16x16 etc. array of pixels) of the camera's imaging sensor. However it is to be appreciated that using parts of the image corresponding to single pixels may also be effective.

The processor 106 is configured to calculate a first value based on TA 210 and T_B 208 for each part of the image. The processor 106 in a preferred embodiment calculates the magnitude of the difference between the two values ΔΤ at step S306. This first value is then compared to a predetermined threshold value Tc based on the detection characteristics of the sensor of the thermographic camera 102, and a level of confidence is assigned to each part of the image based on the comparison as described below. In further embodiments other measures of comparisons of TA 210 and T_B 208 may be used.

If it is determined that ΔΤ is less than Tc for a part of the image at step S308, the imaging device determines that gas detection is unreliable for the location corresponding to that part of the image (i.e. as defined by the pixels for which the measurement is performed) at step S310, and the method continues to figure 4. If it is determined that ΔΤ is greater than or equal to Tc at step S312 the imaging device determines that gas detection is reliable for the location corresponding to that part of the image at step S314, and the method continues to figure 5. For example, Tc may correspond to the detection limit of the camera's sensor. If ΔΤ is determined to be less than Tc in one or more parts of the image, those parts of the image are assigned a low level of confidence as ΔΤ is below the detection limit. Similarly if ΔΤ is determined to be much greater than Tc in one or more parts of the image, those parts of the image are assigned a high level of confidence as ΔΤ is much greater than the detection limit. In further embodiments if ΔΤ is determined to be only slightly greater than Tc, or to be exactly Tc, those parts of the image may be assigned an intermediate level of confidence to indicate that the reading should be treated cautiously as ΔΤ is close to, or at, the detection limit.

Alternatively, the imaging device 100 is configured to perform steps S300 to S310 (or steps S300 to S314, depending on the value of ΔΤ), then return to step S300. In this case no gas detection analysis is performed. When configured to follow these steps, the imaging device 100 may be used to survey one or more locations, and identify areas which might have a low or intermediate level of confidence. It is to be appreciated however that the imaging device 100 could still be used for surveying purposes when if gas detection is being performed.

For example, the imaging device could be used to survey an extended site such as an oil refinery. Certain structures on the site may have surfaces with a certain reflectivity or emissivity, such that ΔΤ for the surfaces is likely to be small, and therefore the detection of gas using known passive thermal imaging techniques may be difficult. The imaging device is used to identify such surfaces and inform the user accordingly. The user may then take appropriate action, for example painting the surfaces to reduce their reflectivity. Figure 4 is a flowchart of the method used by imaging device 100 if it is determined that the level of confidence is such that gas detection for the part of the image is deemed unreliable in step S310.

As the level of confidence may change on a region by region, and indeed a pixel by pixel, basis the methodology described with respect to Figure 4 is only applied to the areas of the thermal image which are deemed unreliable i.e. the areas where ΔΤ is low. At step S400, the processor is configured exclude the part of the image from gas detection analysis as described above. In the present embodiment the processor is further configured to complete the step of causing the display 114 to display an indication that gas detection is deemed unreliable in that part of the image (step not shown).

In such embodiments the pixels which are deemed to correspond to regions of low confidence (i.e. those having a ΔΤ below the threshold limit of the camera) are highlighted to the user when the thermal image is rendered on the display (which is either at the thermal camera or at a remote location such as a control room).

For example where the thermal image is presented to the user, regions/pixels with low confidence and where the presence of gas has been detected are coloured in a different colour so that the user may easily identify such regions. This pixels/regions would therefore be easily identifiable as being "blind" pixels. Accordingly users in charge of safety in, for example a gas processing plant, would therefore be able to act appropriately to an alarm. For example, if the alarm is an area which is known to "blind" then they may await further confirmatory detection of the gas before taking action.

In a preferred embodiment, the system searches for the presence of methane (in the known manner) across the whole image whilst highlighting areas of low confidence as described above. The system is further configured to separately perform the same calculation whilst blocking off/ignoring the pixels which have been identified as being unreliable. By presenting the end user with the results of both calculations the user is able to make the determination as to whether a gas cloud is present in a given region.

At step S402 the imaging device 100 is configured to return to step S300 to repeat the method for a subsequent thermal radiation image generated by the thermographic camera 102.

Returning to the example above, figure 4 corresponds to a situation in which one or more parts of the image has been assigned a low level of confidence. In this case ΔΤ is less than the detection limit of the camera's image sensor, perhaps due to atmospheric conditions or due to the reflectivity of a background structure being high as described above. For these parts of the image, gas detection analysis is considered not to be reliable as the imaging device is operating beyond the detection limit of the camera. Accordingly these parts of the image are excluded from gas detection analysis to prevent potentially erroneous data being provided to the user. Preferably the user is alerted that the imaging device is considered not to able to detect target gas for a location corresponding to the parts of the image. Thus the user is able to take appropriate action as necessary.

Figure 5 outlines the method used by imaging device 100 if it is determined that the level of confidence is such that gas detection for the part of the image is deemed reliable in step S314.

At step S500, the processor is configured include the part of the image when performing gas detection analysis on the image as described above.

At step S502 the imaging device 100 is configured to return to step S300 to repeat the method for a subsequent thermal radiation image generated by the thermographic camera 102. Figure 5 corresponds to a situation in which the level of confidence is either high or intermediate for a part of the image. In this case ΔΤ is either well above the detection limit of the camera's sensor, or is close to or at the detection limit respectively. For one or more parts of the image having a high level of confidence, gas detection analysis proceeds normally. For one or more parts of the image having an intermediate level of confidence, gas detection analysis may proceed as normal, though the imaging device may alert the user that the camera is operating in conditions close to its sensor's detection threshold, and indicate the parts of the image accordingly.

By implementing the method described in figures 3 to 5, the present invention advantageously identifies which parts of a thermal image of a location correspond to areas in which gas detection is likely to be reliable and areas in which it is not. By indicating this on a display (for example a remote display housed in a control room), the user is informed when and where it may be necessary to instigate additional safety measures, for example limiting personnel access to an area in which gas detection has been deemed unreliable.

The processor 106 may be further configured to track the value of ΔΤ for each part of a series of images, and predict the future value of ΔΤ in subsequent images. Thus if one or more parts of the image correspond to a low level of confidence, the processor is able to estimate when conditions will allow for the detection of gas in those parts, and cause the display to indicate this to a user.

The invention as described above may be particularly advantageous when the imaging device is being used to detect leaks of target gas in outdoor locations during adverse weather conditions, for example an oil refinery in foggy conditions. When a dense region of fog enters the camera's field of view, ΔΤ may fall below the detection limit associated with the thermographic camera's detector in one or more parts of the image. The imaging device excludes these parts of the image from gas detection analysis, thus preventing potentially erroneous data to the user. The imaging device provides a visual indication of the affected parts to the user, who is then able to take any appropriate action, such as limiting personnel access to the areas corresponding to the parts of the image with a low confidence level. The device then tracks the value of ΔΤ in each part of the image, and predicts when the fog is likely to have thinned enough to allow gas detection analysis to be performed in that part of the image. An indication of this time is provided to the user via the display, the user then being able to plan appropriate action accordingly. In cases where the fog does not completely dissipate, ΔΤ may by higher than the detection threshold of the camera's sensor, but may still be small in one or more parts of the image. In this case the user is informed that these parts have been assigned an intermediate confidence level, and the user thus knows that conditions may change such that ΔΤ falls below the threshold of the camera's sensor.

When a particular region is identified as having a low or intermediate level of confidence, the user may place the heat source in that particular region. Having measured the temperature of the background radiation emanating from the region, the user may then choose a temperature of radiation to be emitted by the heat source, such that ΔΤ is increased to a point at which the region may be assigned a high level of confidence. Thus the heat source may also be used to enable gas detection in regions of low ΔΤ, in addition to calibrating the imaging device 100.

For example, on measuring ΔΤ for a particular location, the imaging device determines that a certain structure has a value of ΔΤ that is lower than the threshold of the sensor of the camera 102, for example due to the reflectivity or emissivity of the structure. The imaging device indicates this to the user, who then positions the heat source at the location, in front of the structure as seen by the camera. The user then chooses a temperature for the heat source. This temperature may be chosen such that it is greater than the measured ambient temperature by a certain amount, for example 10K. Alternatively the temperature could be based on the previously measured background radiation temperature. The heat source emits radiation at the chosen temperature, effectively changing the temperature measured by the imaging device, such that ΔΤ is now high enough that the structure is assigned a high level of confidence.

Therefore as well as aiding in the calibration of the camera the heat source may proactively improve the camera's ability to detect gas in regions where the confidence level in a detection is low.

In the embodiment above, the threshold value has been described as being based on the detection limit of the camera's image sensor. It is to be appreciated that if the sensor has a plurality of pixels with different detection limits, a different threshold value can be determined for each pixel.

Furthermore, such techniques may be used when commissioning or installing the systems in a given location. For example, if it is known that a certain feature (e.g. a refrigerated structure, a building painted in a paint with a low emissivity) would typically result in a significant portion of the image in the field of view being deemed unreliable due to the low ΔΤ the camera would be positioned so as to minimise the effect. Similarly remedial action, such as repainting a building may be undertaken if necessary.

Claims

1. An imaging device for use in gas detection comprising:

a thermographic camera; and

a sensor for measuring temperature;

the device configured to:

generate a first thermal image of radiation of a first location using the camera; measure an ambient temperature of the location using the sensor; calculate a first value based on a determined temperature for a first portion of the first thermal image and the measured ambient temperature;

compare the first value of the first portion of the first image to a predetermined value indicative of a detection limit of the camera; and

assign a level of confidence to the first portion of the first image based on the comparison.

2. The device of claim 1 further configured to determine a level of confidence for one, or more, further portions of the thermal image based on a comparison.

3. The device of claim 2 further configured to detect the presence of a target gas in one or more portions of the first thermal image.

4. The device of claim 3 wherein the device is further configured, when detecting the presence of the target gas, to exclude one or more portions of the image which have a level of confidence below a predetermined level of confidence.

5. The device of claim 3 or 4, the device is further configured to determine the level of confidence for the one or more portions of the first thermal in which the target gas is detected and in the event that the level of confidence exceeds a predetermined threshold generate signal to indicate the detection of the target gas.

6. The device of claim 3, the device further configured that in the event that the level of confidence is below the predetermined threshold to generate a signal to indicate the detection of the target gas is deemed unreliable.

7. The device of any preceding claim wherein the first value is the magnitude of the difference between the temperature of a first portion of the first image and the measured ambient temperature.

8. The device of any preceding claim further comprising a filter and configured to produce a plurality of thermal images to determine the species of gas detected.

9. The device of any preceding claim to generate a plurality of thermal images.

10. The device of claim 7 configured to identify a first region having a level of confidence lower than the predetermined threshold in successive images.

11. The device of claim 8 further configured to track the position of the first region over the successive images.

12. The device of claim 8 or 9 further configured to predict the position of the first region in subsequent images.

13. The device of any preceding claim further comprising a display, wherein the device is further configured to display a visual image on the display corresponding to the radiation imaged by the camera.

14. The device of claim 12 wherein the device is further configured to indicate on the display regions which have a level of confidence which has been determined to be lower than a predetermined confidence limit.

15. The device of claim 13 wherein the device is configured to identify the regions of low confidence in an image by means of a visual indicator.

16. The device of any preceding claim wherein the level of confidence is based on a chemical composition of a target gas.

17. The device of any preceding claim further comprising a plurality of optical filters.

18. The device of claim 15 wherein each optical filter may be moved between a first position and a second position.

19. The device of claim 16 wherein in the second position, optical filters are disposed in the path of radiation to be imaged by the camera.

20. The device of any preceding claim further comprising wireless communication circuitry.

21. The device of any preceding claim further comprising wired communication circuitry.

22. The device of any preceding claim further comprising a remote heat source located in a field of view of the camera, wherein the heat source is configurable to generate thermal radiation at one or more known temperatures.

23. The device of claim 18 further configured to calibrate the thermographic camera based on the known temperature of the thermal radiation generated by the heat source.

24. A method for determining a level of confidence in a gas detection system, the gas detection system comprising a thermographic camera and a sensor for measuring temperature, the method comprising the steps of:

generating a first thermal image of radiation of a first location using the camera;

measuring an ambient temperature of the location using the sensor; calculating a first value based on a determined temperature for a first portion of the first thermal image and the measured ambient temperature;

comparing the first value of the first portion of the first image to a predetermined value indicative of a detection limit of the camera; and assigning a level of confidence to the first portion of the first image based on the comparison.

25. The method of claim 24 further comprising the step of:

determining a level of confidence for one, or more, further portions of the thermal image based on a comparison.

26. The method of claim 25 further comprising the step of:

detecting the presence of a target gas in one or more portions of the first thermal image.

27. The method of claim 26 further comprising the step of:

excluding the one or more portions of the image which have a level of confidence below a predetermined level of confidence when detecting the presence of the gas in the thermal image.

28. The method of claim 25 or 26, further comprising the step of:

determining the level of confidence for the one or more portions of the first thermal in which the target gas is detected and in the event that the level of confidence exceeds a predetermined threshold generating a signal to indicate the detection of the target gas.

29. The method of any of claims 25 to 28 further comprising the step of:

displaying, on a display, a visual image on the display corresponding to the radiation imaged by the camera.

30. The method of claim 29 further comprising the step of:

indicating on the displayed image regions which have a level of confidence which has been determined to be lower than a predetermined confidence limit.

31. The method of claim 30 wherein the step of indicating comprises identifying the regions of low confidence in an image by means of a visual indicator.