CAMERA BASED IDENTIFICATION OF A CONDITION OF A FLUID COMPONENT Preferred embodiments of the invention relate to a method for determining a condition of a fluid component. Such a method preferably comprises defining a regularity to occur in said exterior of said fluid component during an exchange of fluid between said component and said fluid carrying system; recording during a measuring phase an image sequence of an exterior of said fluid component during which recording the fluid component is fluidic connected to said fluid carrying system, and determining from said recorded image sequence an image feature, if present, expressing said regularity, and if present in the recorded image sequence assigning the condition of the fluid component to be in conformity with said regularity.

BACKGROUND OF THE INVENTION

Fluid components and in particular fluid components exchanging fluid with a fluid carrying system may, by natural causes, fail and become in need for replacement. While different techniques exist for diagnosing such fluid component do exist, e.g., dismantling and examine the different bits and part of the fluid component, such procedures are often cumbersome and labor intensive.

Further, troubleshooting a fluid component e.g., comprised in a larger fluid system as well-functioning or mall-functioning by visual observation of the fluid component may be very difficult or even impossible.

This problem is in particular problematic in relation to expansion vessels, as failure in such vessels often seems periodical at first. Furthermore, failures also look like other failures, such as leaks in other parts of the fluid system, when observed for a couple of hours. The quite long time needed to diagnose correctly is often due to the changes in the fluid system occurring at a rather long timescale, such as hours or even days, and in a central heating system the timescale is also influenced by climate changes. Accordingly, a problem for the service technician during troubleshooting is that he often has to visit the installation without any prior insight.

This often results in that the fluid component if suspected to be malfunctioning is replaced only based on the assumption of being malfunctioning as this may be the cheapest solution, which replacement in some instances solves the problem and in some does not solve the problem. Obviously, this is not an optimal way to make use of resources either being raw materials to produce fluid components, or the labor put into the fixing a problem.

Hence, an improvement method for determining the condition of a fluid component would be advantageous, and in particular a more efficient and/or reliable method for determining the condition of fluid component would be advantageous.

OBJECT OF THE INVENTION

It is an object of the present invention to provide a method and system which at least potentially may increase the likelihood that a fluid component is diagnosed correctly.

It is a further object of the present invention to provide an alternative to the prior art.

In particular, it may be seen as an object of the present invention to provide a method and system which at least potentially may increase the likelihood that a fluid composed is diagnosed correctly in a non-invasive manner, such as without the need for dismantling the fluid component.

SUMMARY OF THE INVENTION

Thus, the above-described objects and several other objects are intended to be obtained in a first aspect of the invention by providing a method for determining a condition of a fluid component,

• said fluid component may comprise a confined space in fluidic connection with a fluid carrying system to exchange fluid between the confined space and the fluid carrying system as a result of a change in pressure and/or temperature change in the fluid carrying system, and wherein the dimension(s) of an exterior of the fluid component changes due to a change in pressure and/or temperature inside the fluid components due to said exchange of fluid, • the method is based on a priori defined a regularity to occur in said exterior of said fluid component during an exchange of fluid between said component and said fluid carrying system; the method may comprise

• recording during a measuring phase an image sequence of an exterior of said fluid component during which recording the fluid component is fluidic connected to said fluid carrying system, and

• determining from said recorded image sequence an image feature, if present, expressing said regularity, and if present in the recorded image sequence assigning the condition of the fluid component to be in conformity with said regularity.

Non-limiting examples on a "fluid component" may be an expansion vessel, a section of a flow pipe, cabinets e.g. encapsulating boilers, a heat exchanger, a hot-water tank, a buffer tank.

Regularity is related to a hypothesis to be proven or non-proven. This may be a faulty (malfunctioning) expansion vessel or well-functioning expansion vessel, which may give rise to a regularity being expressed as a specific movement or non-movement being expressed in the image sequence.

Image feature is used to reference a feature from an image or sequence of images, where the feature expresses the regularity. An image feature may be an edge detection and/or movement of an edge and consequently be a single feature or a composed feature.

In a second aspect the invention relates to a method for motion amplifying exterior movements of a fluid component,

• said fluid component may comprise a confined space in fluidic connection with a fluid carrying system to exchange fluid between the confined space and the fluid carrying system as a result of a change in pressure and/or temperature change in the fluid carrying system, and wherein the dimension(s) of an exterior of the fluid component changes due to a change in pressure and/or temperature inside the fluid components due to said exchange of fluid, the method may comprise

• recording during a measuring phase an image sequence of an exterior of said fluid component during which recording the fluid component is fluidic connected to said fluid carrying system, and

• subjecting a plurality, such all images, of said image sequence to a motion amplification to provide a motion amplified image sequence.

In some embodiments, the condition of the fluid component may be found by a visual inspection of the motion amplified image sequence.

In a third aspect, the invention relates to a system which may comprise

• a digital camera configured to record said image sequence;

• a processor in data communication with said digital camera to receive said image sequence and being configured to carry out the method according to first and/or second aspect

The individual aspects of the present invention may each be combined with any of the other aspects. These and other aspects of the invention will be apparent from the following description with reference to the described embodiments.

The preferred embodiments of invention may be implemented by use of a computer configured to carry out the one or more such as all computable steps and/or one or more such as all steps which can be computerized.

With reference to the first aspect, such a computer implemented method for determining a condition of a fluid component may involve

• said fluid component comprising a confined space in fluidic connection with a fluid carrying system to exchange fluid between the confined space and the fluid carrying system as a result of a change in pressure and/or temperature change in the fluid carrying system, and wherein the dimension(s) of an exterior of the fluid component changes due to a change in pressure and/or temperature inside the fluid components due to said exchange of fluid, • the computer implemented method is based on a priori defined a regularity to occur in said exterior of said fluid component during an exchange of fluid between said component and said fluid carrying system; the computer implemented method comprising

• recording by a digital camera during a measuring phase an image sequence of an exterior of said fluid component during which recording the fluid component is fluidic connected to said fluid carrying system, and

• determining by the computer from said recorded image sequence an image feature, if present, expressing said regularity, and if present in the recorded image sequence assigning by the computer the condition of the fluid component to be in conformity with said regularity.

In a further aspect, the invention relates to a computer program product being adapted to enable a computer system comprising at least one computer having data storage means in connection therewith to carry out one or more computable steps according to the various embodiments of the invention, such as a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out one or more such as all computable steps of the embodiment.

BRIEF DESCRIPTION OF THE FIGURES

The present invention and in particular preferred embodiments thereof will now be described in more details with regard to the accompanying figures. The figures show ways of implementing the present invention and are not to be construed as being limiting to other possible embodiments falling with the scope of the attached claims.

Fig. 1 schematically illustrates a central heating system according to a first embodiment of the invention;

Fig. 2 schematically illustrates two examples on mass expansion quantification metrics (mass expansion refers to volume changes due to temperature changes); in Fig. 2A a preferred embodiment is illustrated where a phase difference detection is based on a binary sum which may be advantageous in connection with less computational power available and Fig. 2B illustrates an embodiment where the phase difference detection is based on entropy calculations which may be advantageously when more computational power is available.

Fig. 3 schematically illustrates phase-based motion amplification according to a preferred embodiment,

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

In the following detailed description of various embodiments of the invention will be disclosed. The detailed description has been made with reference to an application involving an expansion vessel in a central heating system, but the invention is not limited to such application.

Accordingly, the invention is not limited to such a central heating system. In preferred embodiments, the fluid carrying system may be a HVAC system (heating, ventilation and/or air conditioning system) or other systems. Thus, preferred embodiments comprise the fluid component having a confined space in fluidic connection with a fluid carrying system to exchange fluid between the confined space and the fluid carrying system as a result of a change in pressure and/or temperature change in the fluid carrying system, and wherein the dimension(s) of an exterior of the fluid component changes due to a change in pressure and/or temperature inside the fluid components due to said exchange of fluid. The fluid carrying system may be or may comprise a HVAC system, a heat pump, a heat source being powered by electrical power, gas-fired, oil-fired, wood- fired or even coal-fired and/or a cooling system.

Reference is made to Fig. 1 schematically illustrating a central heating system. The central heating system comprising a heat source 2 which may be a boiler, a heat pump or other component which can heat or cool water, or in general a heat carrying fluid, to be circulated in the central heating system. The heat source 2 may be powered by electrical power, gas-fired, oil-fired, wood-fired or even coal- fired. In case where the heat source 2 cools the water, the fluid carrying system may be referred to as a cooling system.

The central heating system comprises besides a number of valve or other fluid flow controlling elements, a utility 3 which may be a radiator, floor-based heating/cooling elements or other heat transferable devices. The heat source 2 and the utility 3 are fluidic connected so that water may be exchanged between the heat source 2 and the utility 3.

During operation of the central heating system, the temperature of the water changes and such changes in temperature result in a volume change of the water. To avoid pressure from increasing in the central heating system, at least to a larger extent, the central heating system comprises an expansion vessel 1 fluidically connected to the central heating system to receive or delivery water in response to the change in volume of the water in the heating system. The expansion vessel 1 may be described as a cavity containing liquid in lower part and gas, typically air or other gaseous fluid, at an upper part. As the upper part of the expansion vessel 1 contains gas, which is considered much more compressible than the water, a change in volume of the water in the heating system will provide a compression of the gas and only minor changes, such as substantial no changes, in the pressure of the water in the heating system. Such function of the expansion vessel is typically referred to as being well-functioning.

However, expansion vessels may become malfunctioning e.g., by a leakage allowing the gas and/or water to flow out from the expansion vessel and into e.g., the surroundings. This is often quite cumbersome to detect e.g., as a typical timescale at which volume expansion of water occurs is quite long - often counted in hours - which requires that a service technician will have to spend long time just monitoring the heating system or frequently return to control the heating system.

However, it has been realized in connection with preferred embodiments, that a well-functioning and malfunctioning expansion vessel provides different exterior footprints, namely if the expansion vessel is well-functioning the upper part and lower part will move out of phase with each other whereas for a malfunctioning expansion vessel, the upper part and lower part will move substantially synchronously with each other. It is noted that the movement considered here are movement arising from a thermal expansion of the material of the wall of the expansion vessel. Accordingly, by monitoring the exterior of the expansion vessel over time, at least a hypothesis as to whether the expansion vessel may be malfunctioning or wellfunctioning may be formulated.

This has been implemented in preferred embodiment of the invention, as a method for determining a condition of a fluid component. In the non-limiting example, the fluid component is an expansion vessel, and the condition is one of malfunctioning or well-functioning. However, other conditions may be considered, such as broken, destroyed or the like.

The fluid component may be disclosed as comprising a confined space in fluidic connection with a fluid carrying system to exchange fluid between the confined space and the fluid carrying system as a result of a change in pressure and/or temperature change in the fluid carrying system, and wherein the dimension(s) of an exterior of the fluid component changes due to a change in pressure and/or temperature inside the fluid components due to said exchange of fluid. In the non-limiting example, the fluid carrying system is a central heating system.

In preferably embodiments, the method typically comprising a step of defining a regularity to occur in said exterior of said fluid component during an exchange of fluid between said fluid component and said fluid carrying system, and the regularity is typically defined by a human or an artificial intelligence routine. Accordingly, the definition of the regularity may be referred to as being defined a priori to carrying out a method according to preferred embodiments of present invention. Such a regularity is typically linked to geometrical changes in the exterior of the fluid component to be observed in a process of testing whether or not a hypothesis is true or false. This may be disclosed by the following two nonlimiting examples:

• Hypothesis: Malfunctioning expansion vessel -> o Regularity: upper section and lower sections moves substantially synchronously with each other.

• Hypothesis: Well-functioning expansion vessel - > o Regularity: upper and lower sections moves out of phase with each other.

Thus, a hypothesis is not limiting to "good" or "bad" but may be defined according to "what to look for".

In preferred embodiments, a measuring phase is carried out during which an image sequence of an exterior of the fluid component is recorded. Such an image sequence typically comprising obtaining a number of individual digital images obtained, typically, at regular intervals. The duration of the image sequence is typically sufficiently long to capture changes which occur at the time scale r (see below) for the system to which the fluid component is connected. For a central heating system, the duration is typically one or more coherent periods of 24 hours but may be less such as 6-12 hours. The longer time periods for a central heating system includes changes occurring, e.g., during shift between night and day and even longer periods may include changes due to climate change. By this, it may be more likely that trends are captured which may be used to consolidate the condition as presented herein. In some embodiments, the equipment used to record image sequences is battery powered and in such embodiments, the recording time is typically 14 days. The frequency at which the images are obtained may also be set in accordance with the time scale, in the sense that the timewise resolution of the image sequence should be sufficient to actually capture changes in the exterior of the fluid component. In some preferred embodiments, the frequency may be time scale divided by a total number of images, such as 500 images, or even 1000 images. In some preferred embodiments, the image sequence frequency (the time between each image) may constant and in other embodiments the image sequence frequency may vary during at least part of the measuring phase and when used during at least part of the calibration phase. By varying the frequency, it may be possible to obtain a high time resolution during time periods where significant changes occur by increasing the frequency in such time periods.

In embodiments including central heating, where the central heating may be based on electrical power, gas, oil, coal or wood combustion, heat pumps or district heating, the heating typically includes heat pulses followed by no heat periods where no heat is produced. A duration of heat pulse plus no heat production, typically varies between 20 minutes to a couple of hours. Depending on the plumping layout, the heating system sets in a steady state mode typically within 10 minutes for both the heat pulse period and the no heat period. The steady state for the heat pulse period may be labelled "active steady state" and the steady state for the no heat production period may be labelled "resting steady state". The steady state, in preferred embodiments, being of interest is typically the steady state (active or resting) when the system is in rest in contrast to the situation where the circulation pump ramps up (to produce a higher volume flow) in response to a higher demand of heat and/or a sudden pressure loss due to a number of heat consuming utilities becomes active. In both situations, images from the resting steady state or active steady state are compared with images obtained after a change in operating pattern recognisable e.g. by pressure change(s), as pressure changes will induce a change in temperature within a system specific time constant T. In preferred embodiments r is selected to be at least 5 minutes such as at least 10 minutes and preferably less than 20 minutes. In a non-limiting example, r is set to be 10 minutes and images may typically be taken every 5 minutes or even more often such as every 2 minutes.

Clearly, during the recording, the fluid component is fluidic connected to said fluid carrying system.

After the image sequence has been provided, an image feature, if present, is determined from the image sequence. The image feature is selected to be a feature comprised in the image sequence that expresses the regularity. As a nonlimiting example, if the regularity is: "upper and lower sections move out of phase" an image feature could be an edge detection of the upper and lower sections of the fluid component including the movements of the upper and lower sections as function of time. Accordingly, an image feature may be a compound feature composed by a number of sub-features, e.g., edge detection and timewise movement. It is to be noted, that the processing of the images needs not to await stopping of the recording as the processing of images may be carried out when an image is available and depending on the computer power available. If the image feature is present in the recorded image sequence and it express said regularity, the condition of the fluid component is typically assigned to be in conformity with said regularity. While this may be used to set the hypothesis to be true - or false if not present - and used, e.g., to determine whether the fluid component, e.g., is malfunctioning or well-functioning, a further evaluation of the hypothesis may be applied as disclosed herein. Such a further evaluation is typically used to strengthen the validity of the hypothesis.

Accordingly, in some preferred embodiments of the invention the regularity is defined as movements of one or more sections of said exterior of the fluid component occurring out of phase relative to movements of one or more other sections of the fluid component.

Accordingly, this may be combined with or used without the regularity being defined as movements of one or more sections of said exterior of the fluid component occurring synchronously relative to movements of one or more other sections of the fluid component.

The movement considered is typically for sections which experiences temperature changes due to exchange of fluid with the fluid carrying system, but other sections may be considered.

As introduced above, preferred embodiments are based on testing a hypothesis and this may symbolically be written as the posterior probability:

P (regularity \image feature) which, if true, provides a value close to one. It is emphasised that while this provides a good approached used in some preferred embodiments, the invention is not limited to determining the posterior probability, as a heuristic approach may also be applied.

Such embodiments may further comprise recording during a calibration a calibration image sequence of the exterior of said fluid component and/or fluid carrying system and determining from said calibration image sequence a steady state image feature. Thus, during the calibration the pressure and/or temperature during is/are substantial constant. The calibration image sequence like the image sequence comprising a number of images obtained with frequency and over a duration typically as detailed above.

Such a steady state image feature may typically be the same feature as the image feature although determined at steady state or substantially steady state. In embodiments involving an expansion vessel, such a steady state image feature may be an edge detection of the upper and lower sections of the fluid component including the movements thereof, although often negligible during steady state, of the upper and lower sections as function of time.

The steady state image feature may be viewed as being obtained during a calibration phase and after the calibration phase, a measuring phase is initiated.

During a measuring phase the image feature is determined. In such embodiments, the image feature is labelled a state image feature and it is evaluated at a state where either the temperature and/or the pressure of the fluid in the fluid carrying system change(s) during the measuring phase.

The pressure and temperature may be evaluated based on an image sequence obtaining images of a pressure sensor and temperature sensor, which images may be separate images or comprised in the recorded images of the exterior of the fluid component.

In embodiments involving an expansion vessel, the state image feature may be an edge detection of the upper and lower sections of the fluid component including the movements thereof.

The condition of the fluid component assigned to be in conformity with said regularity is consolidated to be in conformity if a predefined criteria based on the state image feature and the steady state image feature is fulfilled.

In embodiments involving an expansion vessel the regularity may be considered to be that upper and lower sections of the vessel moves out of phase. To determine the corresponding image feature, the images may be processed by an image amplification process (see below) which amplify motion of, e.g., edges of the vessel. Such processed images may be converted into a binary representation and an edge detection process may be used to locate the edges of the upper and lower sections of the vessel. By such a procedure, movements of the upper and lower sections may be compared to identify whether or not the regularity is represented in the image sequence.

In some embodiments, a so-called "blob-detection" may be used according to which the expansion vessel may be located based on the colour of the expansion vessel, e.g., if the expansion vessel is red, the expansion vessel is detectable by the red colour band of the camera. An image of the expansion vessel, preferably not motion amplification processed, is used to locate an optimal mirror match between top and bottom in the blob by use of block matching. In the binary images the two curves representing the edges of the upper and lower sections of the vessel provide a larger pixel difference sum, when movements are out of phase compared to when movements are synchronously.

Similarly, in embodiments where the entropy is calculated for the images, a similar reason applies as entropy may be considered as an expression for energy composition in the image. Such an entropy may be determined as the energy composition in e.g. a grey scale or black and white images. If a larger change or a binary change between e.g. top and bottom of the fluid component, the entropy between the two overlaid images - or the energy composition - is large (derived pixel entropy). If the top and bottom moves synchronized the entropy - or energy distribution - is small. The quantification "large" and "small" is typically to be carried out by experiments. However, since the entropy may be considered a relative measure, it is often beneficial to determine the quantification large and small during a calibration procedure to identify what may be referred to as a system level.

It is noted that determination of an image feature, such as a phase difference, based on a binary sum calculation may be preferred in situations where the computational power available is low compared to a situation where the computation power available is higher, where the image feature, such as a phase difference, preferably may be determined based on entropy calculations.

Fig. 2 schematically illustrates preferred embodiments of determination of an image feature based on a Blob technique, in which the image feature is movements of the upper section and lower section (in the illustrated embodiments only the upper section is disclosed).

In other embodiments a cross correlation may be used to identify the image feature.

In some preferred embodiments, the predefined criteria is based on a posterior probability determination, and in such embodiments, the criteria can be written as:

P(regularity]state image feature)

H (regularity state image feature) _true

H (regularity] state image feature) _true + H (regularity\steady state image feature)f _aise which may be evaluated as: f (state image feature) P(regularity]image feature) = — - - - r - — - - - - - r f (state image feature) + f (steady state image feature) where f( ) is a function provding a number or vector based on the image feature

Based on this, the assigned condition may be consolidated if the above ratio is larger than 0.75, such as larger than 0.80, such as larger than 0.90 and less than 1.00.

In addition, preferred embodiments may further comprise assigning the condition of the fluid component to be in non-conformity with said regularity if said image feature is not present in the recorded image sequence. Such assignment of non-conformity may be based on a similar approach as disclosed above with regards to consolidating, which involves a calibration phase and a measuring phase to determine a steady state image feature and state image feature. In such embodiments, consolidating the assigned condition of the fluid component to be in non-conformity with said regularity is made if a predefined criteria based on the state image feature and the steady state image feature is fulfilled as outlined above. In preferred embodiments, the predefined criteria may be a ratio between said state image feature and the sum of said steady state image feature and said state image feature is smaller than 0.35, such as smaller than 0.25, such as smaller than 0.15 and larger than 0.00.

In preferred embodiments of the invention, the image feature is determined by calculation an entropy for one or more such as all of the images in the image sequence. Such an entropy may be a single entropy for each of one or more images or it may be an entropy calculated for a number or lines and/or rows in for each of one or more images. Such calculated entropies may then be compared, e.g., picture by picture. If no change, or substantial no change in entropies from image to image is present, it may at least be an indication that no movement of the exterior of the fluid component has taken place, and this may be used to determine whether or not the regularity has been observed.

In the example of an expansion vessel, where a regularity may be that upper and lower section moves out of phase, the entropy is typically calculated individually for an upper and lower section of each image. As the entropy changes with movements in the images, the entropy will - in general - for the upper and lower section of the images progress out of phase, if the regularity is present. Similarly, if the regularity is that the upper and lower sections move synchronously, the entropy in the upper and lower sections of the images will - in general - progress synchronously.

Different methods may be applied to calculate entropy, and in preferred embodiments, the entropy is calculated based on a frequency of a number of grayscale intensity values of grayscale in the each of said images. Also, for the embodiments involving entropy calculation, the condition the fluid component assigned to be in conformity with the regularity, may be consolidated based on calculating a posterior probability. This preferably involves a calibration during which a calibration image sequence is recorded and a measuring phase during which the image feature is determined as disclosed above. Based on this, the posterior probability may be evaluated by:

P(W|F) avg (image Entropy phase after preessure rise) avg(imag eEntropy phase after preessure rise) + avg (image Entropy phase steady state)

Where H refers to the hypothesis on the regularity and E is the entropy. As detailed above, the hypothesis may be said to be true if the value of the posterior probability is close to 1, and the same limits are detailed above may be applied for the posterior probability based on entropy.

The posterior probability may be understood as a feedback to increase the confidence on the measured. The amount of the steady state variable to be included in the evaluation of the probability may typically be determined during a calibration phase during which the state of the fluid component is known or well defined. For the expansion vessel, this may be that the expansion vessel is expected to the filled with water, meaning a gas leaking vessel, well-functioning and/or containing little amount of water, meaning water leaking vessel.

In some preferred embodiments, two optimization experiments may be carried in relations to calibration:

Test calibration 1 (non-limiting example): if the entropy of the image at the upper section is out of phase with the entropy in the lower section, this scenario is attempted to be optimized, with the aim of including the part of the image sequence which optimizes the posterior probability to be closest to 1.0 as steady should not introduce changes in the movement of the exterior. It is often preferred to optimize in an iterative manner to determine the amount of steady state. Test calibration 2 (non-limiting example): The hypothesis to be tested is that if the entropy in the upper and lower sections progress synchronously and the above calculation is used to test this hypothesis.

Motions may be too small in amplitude, below human visual spatio-temporal sensitivity may in some cases be explanatory for condition of the fluid component. In embodiments, where, e.g., the image feature is to be presented to a human, a technical problem is related to how to make the image feature visible to the human.

In addition, the image feature to be determined in the recorded image sequence may be small or even too small in amplitude in relation to the bit-resolution of the camera used to allow for an image analysis procedure determining the image feature correctly.

Thus, in some embodiments, a plurality, such as all images, of said image sequence each are subjected to a motion amplification to provide a motion amplified image sequence. Such a motion amplification may be carried by different methods and one preferred embodiment thereof is disclosed below. It is noted, that while it may be possible not to subject all obtained images to a motion amplification, it may generally be preferred that all or substantially all obtained images are subjected to the motion amplification.

Such motion amplification may be amplification of the magnitude of the motion in the image sequence while maintaining the overall image.

In embodiments where the image feature is or may be difficult to detect by image analysis, it may be preferred to determine the image feature in the motion amplified image sequence.

The motion amplified image sequence may advantageously be displayed on a screen. This will allow, e.g., a service technician to visually identify the image feature and take proper action if need, such as replace the fluid component or conclude that it is well-functioning. As the timescales involved may be quite long, it may be preferred to increase the speed at which the images are displayed, that is preferably displayed with a time difference between each image being less than the time difference between each image when recorded. For instance, if the image sequence spans 24 hours, the motion amplified images may be displayed with a minute or less.

To further assist in rendering the image feature more visible, the motion amplified image sequences may be overlaid the image sequence and displayed on a screen. As above, display may preferably be with a time difference between each image being less than the time difference between each image when recorded.

In some embodiments, the principle applied is based on what is called : "phasebased motion amplification". The processing is based on complex valued steerable pyramids (a mix of spatial decomposition like a Laplacian pyramid and kernels). This may be seen as based on phase based optical flow.

In such amplification, phase variation corresponds to the local motions in spatial sub bands of an image. While an aim is to amplify this subtle motion it may also advantageous to de-noise the phase signal spatially in each sub bands of image, which will result in a motion magnification with less noise. Both complex and real valued steerable pyramids have been found to advantageously be applied to sub octave bandwidth pyramids, which will enable amplification of even the smallest motions. To further increase the spatial support, it extends to sub octave bandwidth filters. An image is decomposed according to spatial scale, orientation, and position. The framework of phase-based motion amplification according to a preferred embodiment is shown in Fig. 3. In preferred embodiments, the limit of a (a is denoted "\alpha" in Fig. 3) for octave-bandwidth (for four orientation) steerable pyramid may be

A a = — 4

A is spatial bandwidth. By this the relation also shows that movements at lower spatial frequencies can be amplified more that movements at higher spatial frequencies. The figure "4" may in principle be arbitrarily selected, although it should be an integer. The figure "4" indicates in this connection that the spatial frequency band is divided into "from left", "from above", "from right" and from "below" with reference to the image framing. Accordingly, in preferred embodiments, the motion amplification comprises for each of the plurality of images:

• a spatial decomposition of the image thereby providing spatial decomposed images, wherein said spatial decomposition is performed at a number of different scales;

• a temporal filtering of each of said spatial decomposed images thereby providing filtered decomposed images, wherein the filtering is a bandpass phase filter for each of said spatial decomposed images;

• an amplification of each of said filtered decomposed images thereby providing amplified filtered decomposed images, and

• reconstructing an image as a motion amplified image by combining the amplified filtered decomposed images into a single image.

This single image is preferably a new image with a similar resolution as the original image being motion amplified. It is noted that motion amplification may be seen as arising from that the subliminal displaced pixels are morphed such as exaggeratedly amplified relatively between each image.

In preferred embodiments, the spatial decomposition may be implemented with a so-called steerable pyramid technique.

In preferred embodiments, the bandpass phase filter may be implemented with so-called "ideal filter" based on Fourier and inverse Fourier and temporal window.

While the above disclosure has been focussed towards preferred embodiments of a method, the invention also relates to a system. Such a system may comprise a digital camera 4 configured to record said image sequence. This camera may be a conventional camera based on a CCD chip or CMOS chip and the resolution may be 1024 pixels by 1024 pixels.

The system typically also comprises a processor 15 in data communication with said digital camera to receive the image sequence. Such a data communication may be a direct data communication and/or a cloud-based solution, in which the image sequence is uploaded to a cloud. In a cloud-based implementation, the processor is configured to be in data communication with the camera via the cloud. An advantage of a cloud-based solution is that a relatively simple installation of a camera can take place in the vicinity of the fluid component and the methods as disclosed herein can be carried out remotely.

To allow the processor to carry out one or more of the methods according to preferred embodiments of the invention, the processor is configured to carry out such methods. Such a configuration may typically comprise uploading to the controller instructions which when executed carries out the methods.

The invention can be implemented by means of hardware, software, firmware or any combination of these. The invention or some of the features thereof can also be implemented as software running on one or more data processors and/or digital signal processors.

The individual elements of an embodiment of the invention may be physically, functionally and logically implemented in any suitable way such as in a single unit, in a plurality of units or as part of separate functional units. The invention may be implemented in a single unit or be both physically and functionally distributed between different units and processors.

Although the present invention has been described in connection with the specified embodiments, it should not be construed as being in any way limited to the presented examples. The scope of the present invention is to be interpreted in the light of the accompanying claim set. In the context of the claims, the terms "comprising" or "comprises" do not exclude other possible elements or steps. Also, the mentioning of references such as "a" or "an" etc. should not be construed as excluding a plurality. The use of reference signs in the claims with respect to elements indicated in the figures shall also not be construed as limiting the scope of the invention. Furthermore, individual features mentioned in different claims, may possibly be advantageously combined, and the mentioning of these features in different claims does not exclude that a combination of features is not possible and advantageous.

List of reference symbols used :

1 Fluid component, such as an expansion vessel

2 Heat source, such as boiler or heat pump 3 Utility, such as radiator

4 Image recorder

5 Draining tube

6 Air bleeding valve 7 Gate valve

8 Backflow valve

9 Manometer

10 Pump

11 Gate valve 12 Gate valve

13 Mixing valve

14 Gate valve

15 Processor