KR20210011473A - Apparatus for detecting airborne mcicrobes - Google Patents

Abstract

The present invention provides an apparatus for measuring floating bacteria in air which can measure floating bacteria in air by using a chaotic wave sensor. According to one embodiment of the present invention, the apparatus for measuring floating bacteria in air comprises: a sampling unit including a storage tank to accommodate a sampling solution therein, an intake air flow path on one side of the storage tank to suck outside air to guide the outside air to the sampling solution, and a discharge air flow path on the other side of the storage tank to discharge air of the storage tank to the outside; a wave source to emit a wave towards the sampling solution of the sampling unit; an image sensor to measure wave speckles created by multiple scattering of the wave in the sampling solution in a time series; and a control part to detect whether microbes exist in the sampling solution based on changes of the measured wave speckles over time.

Description

Airborne bacteria measuring device{APPARATUS FOR DETECTING AIRBORNE MCICROBES}

Embodiments of the present invention relate to an apparatus for measuring airborne bacteria.

In general, as a method of measuring microorganisms floating in the air, there is a method for measuring suspended bacteria. The method for measuring suspended bacteria is a method of forcibly inhaling a certain amount of air, passing through a medium, and culturing microorganisms adsorbed on the medium. There is an advantage in that a result that is very close to the number of bacteria can be obtained, but there is a problem in that the result varies depending on the culture conditions or the measurement location, and the time to obtain the result is very long.

In order to solve the above problems and/or limitations, an object of the present invention is to provide an apparatus capable of measuring airborne bacteria using a chaotic wave sensor.

According to an embodiment of the present invention, a storage tank for accommodating a collection liquid therein, an intake passage for sucking external air to one side of the storage tank and guiding it to the collection liquid, and an air of the storage tank outside the other side of the storage tank A collection unit having an exhaust flow path for discharging, a wave source for irradiating a wave toward the collecting liquid of the collecting unit, and an image sensor for time-sequentially measuring a wave speckle generated by multiple scattering of the wave in the collecting liquid And a control unit configured to detect the presence or absence of microorganisms in the collected liquid based on a change in the measured wave speckle over time.

In an embodiment of the present invention, the control unit may obtain a temporal correlation of the wave speckle and detect the presence of microorganisms in the collection solution based on the obtained temporal correlation. .

In one embodiment of the present invention, the collection unit may further include a collection liquid discharge pipe for discharging the collected liquid after detection, and a collection liquid inlet pipe for introducing a new collection liquid into the storage tank.

In one embodiment of the present invention, a first valve installed in the collecting liquid inlet pipe and selectively opening and closing the collecting liquid inlet pipe, and a first valve installed in the collecting liquid discharge pipe, selectively opening and closing the collecting liquid discharge pipe It further includes 2 valves, and the controller may control the operation of the first valve or the second valve by a preset program.

In one embodiment of the present invention, further comprising a sterilization unit for sterilizing the collected liquid in the storage tank has been detected, the control unit is the second to discharge the collected liquid after the sterilization treatment is completed. The operation of the valve can be controlled.

In an embodiment of the present invention, a filter unit installed in the intake passage may further include a filter unit for filtering substances of a predetermined size or more contained in the external air introduced into the intake passage.

In one embodiment of the present invention, the collection unit further includes a multiple scattering amplification unit configured to amplify the number of multiple scattering in the collection liquid by reflecting at least a part of the wave emitted from the collection liquid to the collection liquid. I can.

In an embodiment of the present invention, the control unit may classify the type or concentration of microorganisms present in the collection liquid based on the change over time of the wave speckle.

In an embodiment of the present invention, the control unit machine learns the classification criteria for microorganisms based on the change over time of the wave speckle measured in the order of the time series, and exists in the collection liquid using the microorganism classification criteria. The type or concentration of the microorganism can be distinguished.

Other aspects, features, and advantages other than those described above will become apparent from the following drawings, claims, and detailed description of the invention.

The apparatus for measuring airborne bacteria according to embodiments of the present invention collects airborne bacteria in a collection liquid, and then detects the presence of microorganisms in the collection liquid using a wave speckle, without using a separate chemical method. Airborne bacteria can be quickly and accurately detected. In addition, the airborne bacteria measurement device obtains a classification standard for classifying the type or concentration of microorganisms using the change in the time correlation of the speckle through machine learning, thereby quickly and accurately determining the type or concentration of airborne bacteria. Can be distinguished. Through this, it is possible to determine the presence or absence of pathogens in the air in the region, and to effectively check regional quarantine information through the network environment.

1 is a diagram schematically showing an apparatus for measuring airborne bacteria according to an embodiment of the present invention.
FIG. 2 is a conceptual diagram illustrating a measurement principle of the airborne bacteria measuring apparatus of FIG. 1.
3 is a view for explaining the principle of a chaotic wave sensor according to an embodiment of the present invention.
4 is a block diagram of the apparatus for measuring airborne bacteria of FIG. 1.
5 is a view for explaining a process of processing a collected liquid after the airborne bacteria measurement apparatus of FIG. 1 measures airborne bacteria.
6 is a diagram illustrating an example of a network environment according to an embodiment of the present invention.
7 is a block diagram schematically showing a server according to an embodiment of the present invention.
8 is a diagram illustrating a method of analyzing a temporal correlation of speckles in a learning unit according to an embodiment of the present invention.
9 is a diagram showing a distribution of standard deviations of light intensity of wave speckles measured over time.
10 is an exemplary diagram of a convolutional neural network according to an embodiment of the present invention.
11 and 12 are diagrams for explaining the convolution operation of FIG. 10.
13 is a graph comparing predicted microbial information (prediction) obtained by measuring airborne bacteria through machine learning according to an embodiment of the present invention and actual microbial information (ground truth).

Hereinafter, the following embodiments will be described in detail with reference to the accompanying drawings, and when describing with reference to the drawings, the same or corresponding components are assigned the same reference numerals, and redundant descriptions thereof will be omitted.

Since the present embodiments can apply various transformations, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. Effects and features of the present embodiments, and a method of achieving them will become apparent with reference to the contents described later in detail together with the drawings. However, the present embodiments are not limited to the embodiments disclosed below and may be implemented in various forms.

In the following embodiments, terms such as first and second are not used in a limiting meaning, but are used for the purpose of distinguishing one component from another component.

In the following embodiments, expressions in the singular include plural expressions, unless the context clearly indicates otherwise.

In the following embodiments, terms such as include or have means that the features or elements described in the specification are present, and do not preclude the possibility that one or more other features or components may be added.

In the following embodiments, when a part, such as a unit, a region, or a component, is on or on another part, not only is it directly above the other part, but also another unit, region, component, etc. is interposed therebetween. Includes cases.

In the following examples, terms such as connect or combine do not necessarily mean direct and/or fixed connection or combination of two members, unless the context clearly means differently, and that another member is interposed between the two members. It is not to exclude.

It means that a feature or component described in the specification is present, and does not preclude the possibility that one or more other features or components may be added.

In the drawings, components may be exaggerated or reduced in size for convenience of description. For example, the size and thickness of each component shown in the drawings are arbitrarily shown for convenience of description, and thus the following embodiments are not necessarily limited to those shown.

1 is a diagram schematically showing an apparatus 10 for measuring airborne bacteria according to an embodiment of the present invention, and FIG. 2 is for explaining the measurement principle of the apparatus 10 for measuring airborne bacteria in FIG. It is a conceptual diagram. 3 is a view for explaining the principle of the chaotic wave sensor according to an embodiment of the present invention, Figure 4 is a block diagram of the airborne bacteria measuring apparatus 10 of Figure 1, Figure 5 is the air of Figure 1 It is a diagram for explaining a process of processing the collected liquid after the airborne bacteria measurement device 10 measures airborne bacteria.

In the case of a conventional air sampler, in order to measure airborne bacteria, air containing airborne bacteria is sucked and then spread in a medium to obtain a measurable amount. It was difficult to detect airborne bacteria quickly due to the long time.

The present invention is to solve the above problems, the airborne bacteria measuring apparatus 10 according to an embodiment of the present invention is to contact the air floating microorganisms such as bacteria and bacteria with a collection liquid to It is characterized in that airborne bacteria are detected in a short time by collecting them in the collecting liquid and detecting them using a wave speckle.

First, referring to FIGS. 1 and 2, the airborne bacteria measurement apparatus 10 according to an embodiment of the present invention includes a collection unit 1000, a wave source 100, an image sensor 300, and a control unit 400. ) Can be included.

The collection unit 1000 performs a function of accommodating the collection liquid and collecting airborne bacteria from the inhaled air. Specifically, the collection unit 1000 includes a storage tank 1001 for accommodating the collection liquid W therein, and an intake passage 1101 for inhaling external air to one side of the storage tank 1001 to guide the collection liquid W. , An exhaust passage 1102 for discharging the air from the storage tank 1001 to the outside may be provided on the other side of the storage tank 1001.

Here, the collection liquid W may be any type of sample capable of collecting airborne bacteria through contact with airborne bacteria. In other words, the collection liquid W may be a liquid or a gel sample. However, in the present invention, in order to accurately detect the presence or concentration of airborne bacteria, the collection liquid (W) is prepared in a state that does not contain impurities such as microorganisms before contact with air, or a chaotic wave sensor to be described later is used. The concentration of impurities including microorganisms in the collection liquid W can be measured in advance and stored in advance before contacting with air.

As an embodiment, the collection liquid W may be made of a liquid, and at this time, it may include a culture material for culturing microorganisms. The collection liquid (W) may contain various microbial culture materials, and is described in, for example, "Manual of Methods for General Bacteriology" by the American Society for Bacteriology, Washington D.C., USA, 1981. These media contain various carbon sources, nitrogen sources and trace element components. Carbon sources include carbohydrates such as glucose, lactose, sucrose, fructose, maltose, starch and fiber; Fats such as soybean oil, sunflower oil, castor oil and coconut oil; Fatty acids such as palmitic acid, stearic acid and linoleic acid; Alcohol such as glycerol and ethanol and organic acids such as acetic acid may be included, and these carbon sources may be used alone or in combination, but are not limited thereto. Nitrogen sources include organic nitrogen sources and urea such as peptone, yeast extract, gravy, malt extract, corn steep liquor (CSL) and bean flour, Inorganic nitrogen sources such as ammonium sulfate, ammonium chloride, ammonium phosphate, ammonium carbonate and ammonium nitrate may be included, and these nitrogen sources may be used alone or in combination, but are not limited thereto. The medium may additionally contain potassium dihydrogen phosphate, potassium dihydrogen phosphate, and corresponding sodium-containing salts as a phosphoric acid source, but is not limited thereto. . Further, the medium may contain a metal such as magnesium sulfate or iron sulfate, and amino acids, vitamins, and suitable precursors may be added.

In addition, in order to maintain the aerobic condition of the collecting liquid W, oxygen or a gas containing oxygen (eg, air) may be injected into the collecting liquid W into the collecting liquid W. The temperature of the collecting liquid W may be generally 20-45°C, specifically 25-40°C.

At least part of the storage tank 1001 may be made of a material through which waves can pass. For example, the storage tank 1001 may be formed of a material such as glass at least the area where the wave is incident or the area where the wave is emitted. In another embodiment, when the storage tank 1001 is formed of a material through which the wave cannot penetrate, for example, a metal material, a through hole may be formed corresponding to an area where the wave is incident and the area where the wave is emitted. In this case, the storage tank 1001 may further include a cover part (not shown) disposed in the through hole. The cover part (not shown) may be made of a transparent or translucent material so that waves can be transmitted. For example, the cover part (not shown) may be made of a glass or plastic material, may be formed in a flexible plate shape, or may be made of a film. As another embodiment, the cover part (not shown) may be formed in a form that fills the inside of the through hole rather than a form disposed on one side of the through hole.

On the other hand, the collection unit 1000 reflects at least a part of the wave emitted from the collection liquid W into the collection liquid W to amplify the number of multiple scattering in the collection liquid W. It may further include a unit (MS). For example, the multiple scattering amplification unit MS enters the storage tank 1001 and is scattered by suspended bacteria in the collection liquid W, and then at least a part of the wave emitted from the storage tank 1001 is again collected. ) Can be scattered. The scattered wave is again scattered and emitted by the suspended bacteria in the collection liquid W, and through this process, the number of multiple scattering in the collection liquid W can be increased.

As an embodiment, the multiple scattering amplification unit MS is disposed between the storage tank 1001 and the wave source 100, and is disposed between the storage tank 1001 and the image sensor 300 to perform multiple scattering from the collection liquid (W). And at least a part of the emitted wave can be reflected. In other words, the multiple scattering amplification unit (MS) allows the waves emitted by multiple scattering from the collection solution (W) to reciprocate at least once or more in the space between the collection solution (W) and the multiple scattering amplification unit (MS). I can. The multiple scattering amplification unit MS may be formed by including a multiple scattering material, and for example, the multiple scattering material may include titanium oxide (TiO 2 ).

As another embodiment, the collection unit 1000 may form a multiple scattering amplification region including a multiple scattering material in at least a part of the storage tank 1001. For example, a multi-scattering amplification region may be formed by coating with a multi-scattering material on at least a region of the storage tank 1001 corresponding to a space in which the collecting liquid W is accommodated. However, the technical idea of the present invention is not limited thereto, and a multiple scattering amplification area may be disposed in the entire area of the storage tank 1001 except for the incident portion and the emission portion of the wave.

On the other hand, the intake passage 1101 is disposed on one side of the storage tank 1001, and performs a function of inhaling external air and guiding it to the collected liquid W, and the exhaust passage 1102 is disposed on the other side of the storage tank to be 1001) air can be discharged to the outside. As shown, the intake passage 1101 may be formed in a form in which one end is immersed in the collecting liquid W so that air can directly contact the collecting liquid W. However, the present invention is not limited thereto, and of course, the intake passage 1101 may be formed in any form capable of guiding the outside air with the collecting liquid W.

Meanwhile, a filter unit 1400 may be installed in the intake passage 1101. The filter unit 1400 may perform a function of filtering a substance having a predetermined size or more contained in the external air introduced into the intake passage 1101.

One end of the exhaust passage 1102 may be disposed on the receiving space of the storage tank 1001 except for the collection liquid W so that the air in the storage tank 1001 can be discharged to the outside. The external air is guided to the collection liquid (W) through the intake flow path (1101), and then the residual air excluding the collected airborne bacteria escapes from the collection liquid (W), and the exhaust flow path (1102) discharges such air to the outside. It performs the function to let you know. Although not shown, the collection unit 1000 may connect the exhaust passage 1102 with a means such as a pneumatic pump or a compressor to discharge the air inside the storage tank 1001 to the outside. In addition, by generating a pressure difference in the storage tank 1001 through the above-described process, the collection unit 1000 may suck air into the storage tank 1001.

The collection unit 1000 includes a collection liquid inlet pipe 1201 for introducing the collected liquid W stored externally into the storage tank 1001, and a collection liquid that discharges the detected collection liquid W from the storage tank 1001. It may further include a discharge pipe (1202). The collected liquid inlet pipe 1201 and the collected liquid discharge pipe 1202 may be installed under the storage tank 1001 as shown in the drawing to facilitate inflow and discharge of the collected liquid W.

Meanwhile, the wave source 100 may irradiate a wave toward the collecting liquid W of the collecting unit 1000. The wave source 100 may be applied to any type of source device capable of generating a wave, and may be, for example, a laser capable of irradiating light of a specific wavelength band. The present invention is not limited to the type of wave source, but, for convenience of explanation, the following description will be made focusing on the case of a laser.

For example, a laser having good coherence may be used as the wave source 100 in order to form a wave speckle in the collecting liquid W. In this case, as the spectral bandwidth of the wave source that determines the coherence of the laser wave source decreases, the measurement accuracy may increase. That is, as the coherence length increases, measurement accuracy may increase. Accordingly, laser light having a spectral bandwidth of the wave source less than a predefined reference bandwidth may be used as the wave source 100, and measurement accuracy may increase as the wave source is shorter than the reference bandwidth. For example, the spectral bandwidth of the wave source may be set so that the condition of Equation 1 below is maintained.

[Equation 1]

According to Equation 1, in order to measure the pattern change of the wave speckle, when light is irradiated into the collecting liquid W at every reference time, the spectral bandwidth of the wave source 100 may be maintained less than 5 nm.

The image sensor 300 may measure a wave speckle generated by multiple scattering within the collection liquid W in time series. In other words, the image sensor 300 is disposed on the emission path of the wave, and may acquire a plurality of images by photographing the wave emitted from the collection liquid W in chronological order. The image sensor 300 may include a sensing means corresponding to the type of the wave source 100. For example, when using a light source in the visible wavelength band, a CCD camera that is a photographing device for photographing an image Can be used.

Here, each of the plurality of images may include speckle information generated by multiple scattering by airborne bacteria due to a wave incident on the collecting liquid W. In other words, the image sensor 300 may detect a wave speckle generated by multiple scattering of the irradiated wave within the collection liquid W at a predetermined time point. Here, the term "time" means any one moment in the continuous flow of time, and the time points may be set in advance at the same time interval, but are not necessarily limited thereto, and are set in advance at an arbitrary time interval. It could be.

The image sensor 300 detects a first image including first speckle information at least at a first view point, captures a second image including second speckle information at a second view point, and the controller 400 Can be provided as Meanwhile, the first viewpoint and the second viewpoint are only one example selected for convenience of description, and the image sensor 300 may capture a plurality of images at a plurality of viewpoints greater than the first viewpoint and the second viewpoint. The image sensor 300 includes a polarizer 305 in a path through which a wave is emitted, thereby maximizing interference efficiency for speckle formation and removing unnecessary external reflected light.

The projection sensor 300 may be disposed at a position spaced apart from the collection unit 1000 by a predetermined distance so that the size d of one pixel of the image sensor is smaller than or equal to the grain size of the speckle pattern. For example, the image sensor 300 may be disposed on a path through which the emission wave passes so as to satisfy the condition of Equation 2 below.

[Equation 2]

As shown in Equation 2, the size d of one pixel of the image sensor 300 must be less than or equal to the grain size of the speckle pattern, but if the size of the pixel is too small, undersampling occurs. Difficulties may exist in utilizing the pixel resolution. Accordingly, in order to achieve an effective signal to noise ratio (SNR), the image sensor 300 may be arranged such that up to 5 or less pixels are located in a speckle grain size.

The controller 400 may detect the presence or absence of suspended bacteria, that is, microorganisms in the collected liquid W based on the change of the measured wave speckle over time. As an embodiment, the controller 400 may acquire a temporal correlation of the wave speckle and detect whether or not a microorganism in the collection liquid W is present based on the acquired temporal correlation.

Specifically, the control unit 400 uses the difference between the first image information of the first wave speckle measured at the first viewpoint and the second image information of the second wave speckle detected at a second viewpoint different from the first viewpoint. Thus, the presence or absence of microorganisms (M) can be detected. Here, the first image information and the second image information may be at least one of wave speckle pattern information and wave intensity information.

On the other hand, an embodiment of the present invention does not use only the difference between the first image information at the first viewpoint and the second image information at the second viewpoint, but expands this to provide an image of a plurality of wave speckles at a plurality of viewpoints. Information is available. The control unit 400 may calculate a time correlation coefficient between images using image information of the wave speckle generated for each of a plurality of preset viewpoints, and the presence or absence of microorganisms in the collection liquid based on the time correlation coefficient, Furthermore, it can detect the presence of airborne bacteria. The time correlation of the detected wave speckle image can be calculated using Equation 3 below.

[Equation 3]

은 시간 상관 관계 계수,

은 표준화된 빛 세기, (x,y)는 카메라의 픽셀 좌표, t는 측정된 시간, T는 총 측정 시간,

는 타임래그(time lag)를 나타낸다.In Equation 3

Is the time correlation coefficient,

Is the normalized light intensity, (x,y) is the pixel coordinates of the camera, t is the measured time, T is the total measured time,

Represents a time lag.

The time correlation coefficient may be calculated according to Equation 3, and as an example, the presence of microorganisms may be detected through an analysis in which the time correlation coefficient falls below a preset reference value. Specifically, the presence of microorganisms may be detected as the time correlation coefficient falls below a reference value beyond a preset error range.

Hereinafter, with reference to Figure 3, the principle of the airborne bacteria measuring apparatus 10 of the present invention will be described.

In the case of a material having a homogeneous internal refractive index such as glass, refraction occurs in a certain direction when irradiated with light. However, when a coherent light such as a laser is irradiated onto an object having a non-homogeneous internal refractive index, very complex multiple scattering occurs inside the material.

Referring to FIG. 3, some of the waves scattered in a complex path through multiple scattering among light or waves irradiated from a wave source (hereinafter, referred to as waves for simplicity) pass through the inspection target surface. Waves passing through several points on the inspection target surface cause constructive or destructive interference with each other, and constructive/destructive interference of these waves generates grain-shaped patterns (speckles). do.

In the present specification, waves scattered through such a complex path are referred to as "chaotic waves", and chaotic waves can be detected through speckle information.

Again, the left side of FIG. 3 is a view showing when a stable medium is irradiated with a laser. When a stable medium with no movement of the internal material is irradiated with interference light (eg, laser), a stable speckle pattern without change can be observed.

However, as shown in the right diagram of FIG. 3, when an unstable medium having movement among internal constituent materials such as bacteria is contained therein, the speckle pattern changes.

In other words, the light path may change in real time due to microscopic biological activities of living things (eg, intracellular movement, movement of microorganisms, movement of ticks, etc.). Since the speckle pattern is a phenomenon that occurs due to the interference of waves, a change in a minute light path may cause a change in the speckle pattern. In particular, the change in the minute light path is expressed by a high signal-to-noise ratio due to the characteristics of the data measured by the embodiments of the present invention, which is an image interfered by a light source of a narrow bandwidth, and the signal is caused by multiple Because it is affected times. Accordingly, by measuring the temporal change of the speckle pattern, it is possible to quickly measure the movement of the organism. In this way, when the change of the speckle pattern over time is measured, the existence and concentration of an organism can be known, and furthermore, the type of the organism can also be known.

4 and 5, the apparatus 10 for measuring airborne bacteria according to an embodiment of the present invention may further include a sterilization unit 500.

The sterilization unit 500 may perform a sterilization treatment of the collected liquid W in the storage tank 1001. The sterilization unit 500 may be any means capable of removing microorganisms in the collection liquid (W). For example, the sterilization unit 500 may include at least one of an ultraviolet (UV) lamp or a laser having a predetermined or higher output. Alternatively, the sterilization unit 500 may be sterilized using electrolysis. On the other hand, as another embodiment, the sterilization unit 500 does not sterilize the collection liquid W in the storage tank 1001, but is disposed on the discharge path of the collection liquid discharge pipe 1202 to be discharged. Of course, it can also be sterilized.

On the other hand, the airborne bacteria measuring device 10 is installed in the collected liquid inlet pipe 1201 to selectively open and close the collected liquid inlet pipe 1201, and installed in the collected liquid discharge pipe 1202. It may further include a second valve 1302 for selectively opening and closing the collected liquid discharge pipe 1202. In this case, the controller 400 may control the operation of the first valve 1301 or the second valve 1302 by a preset program.

Specifically, referring to FIG. 5A, the airborne bacteria measurement apparatus 10 may collect the airborne bacteria M in the collection liquid W using the collection unit 1000. As described above, the airborne bacteria measuring device 10 transfers the collected liquid W managed so that the collected liquid W does not contain microorganisms before inhaling the airborne bacteria M into the storage tank 1001 immediately before detection. The concentration of impurities including microorganisms may be stored by introducing inflow or by measuring the collected liquid W accommodated in the storage tank 1001 in advance. When the collected liquid (W) stored externally is introduced into the storage tank (1001), the controller 400 may control the first valve 1301 to open, and after receiving a predetermined amount of the collected liquid (W), the first The valve 1301 can be controlled to be closed. Thereafter, the airborne bacteria measurement apparatus 10 may collect the airborne bacteria (M) contained in the external air by contacting the suctioned external air with the collecting liquid (W).

Referring to FIG. 5B, the airborne bacteria measurement device 10 can detect the presence or concentration of microorganisms (M) in real-time by investigating the waves in the collected liquid (W). However, as another embodiment, it may be detected after culturing the microorganisms (M) in the collection liquid (W) for a predetermined time. If, after culturing the microorganisms (M) in the collection solution (W) for a certain period of time, the measurement is performed, the airborne bacteria measuring device 10 is The actual concentration of suspended bacteria can be estimated.

Referring to (c) of FIG. 5, the collection liquid W, which has been detected as described above, may be sterilized by the sterilization unit 500. Through this, the airborne bacteria measurement device 10 not only minimizes the generation of pollutants by discharging the collected liquid W after sterilization treatment, but more ultimately, by removing microorganisms in the storage tank 1001, The accuracy of the phosphorus detection process can be improved.

Referring to FIG. 5D, when the sterilization treatment is completed, the controller 400 may control the second valve 1302 to open to discharge the collected liquid W in the storage tank 1001 to the outside. By repeating the above process, the airborne bacteria measuring apparatus 10 can repeatedly measure airborne bacteria, and these data can be provided to an external server and used as big data.

Hereinafter, with reference to the drawings, a network environment including the apparatus 10 for measuring airborne bacteria having the above-described configuration will be described, and at this time, not only detection of airborne bacteria, but also concentration or type classification through machine learning. Let's explain how.

6 is a diagram illustrating an example of a network environment according to an embodiment of the present invention.

The network environment 1 of FIG. 6 shows an example including one or more airborne airborne bacteria measuring devices 10, a user terminal 20, a server 30, and a network 40. 6 is an example for explaining the invention, and the number of terminals or servers of a user is not limited as shown in FIG. 6.

One or more airborne bacteria measuring apparatuses 10 may be provided, and are disposed in different regions to detect airborne bacteria in each region through the above-described detection process. The airborne bacteria measuring device 10 may be provided to the server 30 or the user terminal 20 using the network 40 which is a communication network. Using the data provided in this way, the server 30 may generate airborne bacteria information according to regions, more specifically, quarantine information related to airborne bacteria, together with a map, and provide it to the user terminal 20.

The user terminal 20 may be a fixed terminal 22 implemented as a computer device or a mobile terminal 21. The user terminal 20 may be a terminal of an administrator who controls the server 30. Examples of the user terminal 20 include a smart phone, a mobile phone, a navigation system, a computer, a notebook computer, a digital broadcasting terminal, a PDA (Personal Digital Assistants), a PMP (Portable Multimedia Player), a tablet PC, and the like. For example, the user terminal 1 21 may communicate with the other user terminal 22 and/or the server 30 through the communication network 40 using a wireless or wired communication method.

The communication method is not limited, and not only a communication method using a communication network (for example, a mobile communication network, a wired Internet, a wireless Internet, a broadcasting network), but also a short-range wireless communication between devices may be included. For example, the network 40 includes a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), and a broadband network (BBN). , Internet, and the like. In addition, the network 40 may include any one or more of a network topology including a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree or a hierarchical network, etc. Not limited.

The server 30 is a computer device or a plurality of computer devices that communicates with the user terminal 20 or the airborne bacteria measurement device 10 and the network 40 to provide commands, codes, files, contents, services, etc. Can be implemented as

The airborne bacteria measuring apparatus 10 according to an embodiment of the present invention collects airborne bacteria through an independent configuration, measures the wave speckle emitted from the collected liquid, and then waves by the control unit 400 It is possible to learn the classification criteria for microorganisms using the speckle, and perform a function of detecting suspended bacteria in the collected liquid using the learned classification criteria for microorganisms.

However, the technical idea of the present invention is not limited thereto, and the airborne bacteria measurement device 10 transmits detection data to the external server 30, and the external server 30 uses the transmitted data to be After machine learning the classification criteria, the learned algorithm can be provided to the airborne bacteria measuring device 10. The airborne airborne bacteria measurement apparatus 10 may detect the presence of airborne airborne bacteria or classify a concentration or type by using the algorithm provided as described above and data on the newly measured wave speckle. Hereinafter, for convenience of explanation, a description will be made focusing on the case of machine learning the microorganism classification criteria in the server 30.

7 is a block diagram schematically showing the server 30 according to an embodiment of the present invention.

The server 30 may correspond to at least one or more processors, or may include at least one or more processors. Accordingly, the server 30 may be driven in a form included in a hardware device such as a microprocessor or a general-purpose computer system. Here, the'processor' may refer to a data processing device embedded in hardware having a circuit physically structured to perform a function represented by a code or instruction included in a program. As an example of such a data processing device built into the hardware, a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, and an application-specific integrated (ASIC) Circuit) and processing devices such as a Field Programmable Gate Array (FPGA) may be covered, but the scope of the present invention is not limited thereto.

The server 30 shown in FIG. 7 shows only the components related to this embodiment in order to prevent the features of the embodiment from being blurred. Accordingly, it can be understood by those of ordinary skill in the art related to the present embodiment that other general-purpose components may be further included in addition to the components shown in FIG. 3.

Referring to FIG. 7, the server 30 according to an embodiment of the present invention may include an input/output interface 31, a receiver 32, a processor 33, and a memory 34.

The memory 34 is a computer-readable recording medium, and may include a permanent mass storage device such as a random access memory (RAM), read only memory (ROM), and a disk drive.

The input/output interface 31 may be a means for an interface with an input/output device. For example, the input device may include a device such as a keyboard or a mouse, and the output device may include a device such as a display for displaying a communication session of an application. As another example, the input/output interface 31 may be a means for interfacing with a device in which functions for input and output are integrated into one, such as a touch screen.

The receiver 32 may receive a plurality of images from the airborne bacteria measuring apparatus 10. In this case, the plurality of images may be images obtained by photographing waves emitted from the collected liquid W by the image sensor 300 of the airborne bacteria measuring apparatus 10 in chronological order. That is, the receiving unit 32 may function as a communication module using wired or wireless communication and may receive a plurality of images. At this time, the receiving unit 32 may provide a function for communication between the user terminal 1 (21) and the server 30 through the network 40, and other user terminals (for example, user terminal 2 (22)) or other It can provide a function to communicate with the server.

The processor 33 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input/output operations. The instructions may be provided to the processor 33 by the memory 34 or the receiving unit 32. For example, the processor 33 may be configured to execute an instruction received according to a program code stored in a recording device such as the memory 34. The processor 33 may include a learning unit 331, a detection unit 332 and a determination unit 333.

The detection unit 322 may extract a feature of a change over time from a plurality of training images received by the reception unit 32. Here, the plurality of learning images are images continuously photographed at time intervals, and time information is included between the plurality of learning images photographed in chronological order. The detector 322 may extract a feature of change over time from the plurality of learning images.

The learning unit 331 may learn a microorganism classification standard for classifying the type or concentration of microorganisms present in the collection liquid W based on the extracted features. The learning unit 331 learns the classification criteria for microorganisms based on deep learning, and deep learning is a key among high-level abstractions, large amounts of data, or complex data through a combination of several nonlinear transformation methods. It is defined as a set of machine learning algorithms that attempt to summarize basic content or function. The learning unit 331 includes deep neural networks (DNN), convolutional neural networks (CNN), Reccurent Neural Networks (RNN), and deep trust neural networks among models of deep learning. Networks, DBN) may be used.

As an embodiment, the learning unit 331 may machine learning a classification criterion based on a temporal correlation of the received plurality of training images.

As described above, the plurality of learning images may include information on wave speckles generated by multiple scattering from the microbial path M in the collection liquid W. As described above with reference to FIG. 3, when suspended bacteria are present in the collection liquid W, the speckle may change over time due to the biological activity of the microorganism. In addition, since the change of the speckle over time varies according to the type or concentration of the microorganism, the learning unit 331 learns the microorganism classification criteria for classifying the type or concentration of the microorganism using the change over time of the speckle. can do.

8 is a view for explaining a method of analyzing the temporal correlation of speckles in the learning unit 331 according to an embodiment of the present invention.

Referring to FIG. 8, the learning unit 331 may calculate a time correlation coefficient between images as shown in Equation 3, using image information of speckles generated at a plurality of preset viewpoints, and the time correlation It is possible to learn the classification criteria for microorganisms based on the coefficient. As shown in Fig. 8, as the concentration of microorganisms increases, the time correlation coefficient falls below the reference value becomes shorter, and by using this, the concentration of microorganisms can be analyzed through the slope value of the graphic representing the time correlation coefficient. have. The learning unit 331 may learn a microbial classification criterion for classifying the concentration of microbes by using the analysis of the slope value of the time correlation coefficient.

9 is a diagram showing a distribution of standard deviations of light intensity of wave speckles measured over time.

Referring to FIG. 9, the learning unit 331 may calculate a standard deviation of the intensity of light of a speckle pattern by using a plurality of training images measured for each reference time. As bacteria and microorganisms present in the sample continuously move, constructive interference and destructive interference may change in response to the movement. At this time, as constructive interference and destructive interference change, the degree of light intensity may change. The learning unit 331 may analyze a place where bacteria and microorganisms are in a sample by obtaining a standard deviation indicating a degree of change in light intensity, and learn a distribution map of the bacteria and microorganisms.

For example, the learning unit 331 may calculate a standard deviation of light intensity over time of a speckle pattern detected from each of the plurality of training images. The standard deviation of the light intensity according to the time of the speckle may be calculated based on Equation 4 below.

[Equation 4]

: 시간에 따른 평균 빛 세기를 나타낼 수 있다.In Equation 4, S: standard deviation, (x,y): camera pixel coordinates, T: total measurement time, t: measurement time, It: light intensity measured at time t,

: It can show the average light intensity over time.

Since the reinforcing and destructive interference patterns vary according to the movement of bacteria and microorganisms, and the standard deviation value calculated based on Equation 4 is different, the concentration of bacteria and microorganisms can be measured based on this. The learning unit 331 may learn the classification criteria based on a linear relationship between the size of the standard deviation value of the light intensity of the speckle pattern and the concentration of bacteria and microorganisms.

Hereinafter, a case in which the learning unit 331 learns the classification criteria using a convolutional neural network (CNN) will be mainly described.

Here, the convolutional neural network (CNN) is a type of multilayer perceptrons designed to use minimal pre-processing. The convolutional neural network (CNN) includes a convolution layer that performs convolution on input data, and further includes a subsampling layer that performs subsampling on an image, and extracts a feature map from the data. can do. Here, the subsampling layer is a layer that increases a contrast between neighboring data and reduces the amount of data to be processed, and may use max pooling, average pooling, or the like.

10 is an exemplary diagram of a convolutional neural network according to an embodiment of the present invention, and FIGS. 11 and 12 are diagrams for explaining the convolution operation of FIG. 10.

10 to 12, a convolutional neural network (CNN, 2510) used in the learning unit 331 according to an embodiment of the present invention may include a plurality of convolutional layers A1. The learning unit 331 may generate an output by performing a convolution operation between kernels of convolution layers and inputs.

The input of the convolutional layer is data used as an input of the corresponding convolutional layer, and includes at least one input feature map corresponding to initial input data or an output generated by a previous layer. For example, the input of the convolution layer 1 (A11) shown in FIG. 8 is a plurality of images 2501 that are the first inputs of the convolutional neural network 2510, and the input of the convolutional layer 2 (A12) is convolution. It may be the output 2511 of the layer 1 (A11).

The input 2501 of the convolutional neural network 2510 is a plurality of C images 2501 received by the receiver 32, and a time correlation may be established between each image as an input feature map. Each image, that is, each input feature map may have a plurality of pixels having a preset width W and a height H. Since there are C input feature maps, the size of the input 2501 may be expressed as W x H x C. The learning unit 331 may perform a convolution operation corresponding to the input 2501 using one kernel corresponding to the convolution layer A1.

At least one kernel of the convolutional layer A1 is data employed for a convolution operation corresponding to the convolutional layer A1, and may be defined, for example, based on inputs and outputs of the convolutional layer. . At least one kernel may be designed for each convolutional layer A1 constituting the convolutional neural network 2510, and at least one kernel corresponding to each convolutional layer may be referred to as a kernel set S600.

Here, the kernel set S600 may include kernels corresponding to the output channels D. For example, in order to obtain a desired output of the convolutional layer A1, a kernel set S600 of a corresponding convolutional layer may be defined so that an input of a corresponding convolutional layer and a convolution operation are performed. The output of the convolutional layer A1 is data obtained from a result of a convolution operation between an input of a corresponding convolutional layer and a kernel set, and includes at least one output feature map and may be used as an input of a next layer.

The learning unit 331 may generate an output 2511 by performing a convolution operation between the kernel set S600 corresponding to the convolution layer A1 and the input 2501. The output 2511 of the convolutional layer A1 includes output feature maps 25111 corresponding to D output channels, and the size of each output feature map 25111 may be W x H. Here, the width, height, and number (depth) of the output 2511 are W, H, and D, respectively, and the size of the output 2511 may be expressed as W x H x D.

For example, the kernel set S600 corresponding to the convolution layer A1 may include convolution kernels corresponding to the number of D output channels. The learning unit 331 may generate output feature maps 25111 corresponding to the D output channels based on calculation results between the input 2501 and kernels corresponding to the D output channels.

The learning unit 331 includes a plurality of convolutional layers A1, and may include 4 to 7 convolutional layers A1 as an embodiment. For example, as shown in the figure, five convolutional layers A11, A12, A13, A14, and A15 may be included. Through the plurality of convolutional layers A11, A12, A13, A14, and A15, the learning unit 331 can increase a learning capacity capable of implying a complex nonlinear relationship. However, the present invention is not limited thereto, and it goes without saying that learning may be performed using more convolutional layers A1.

Meanwhile, each of the convolutional layers A1 may include an activation function. The activation function is applied to each layer to perform a function of making each input have a complex non-linear relationship. As the activation function, a sigmoid function (Sigmoid), a tanh function (tanh), a rectified linear unit (ReLU), and a leaky relu (Leacky ReLU) that can convert an input into a normalized output can be used. .

The learning unit 331 according to an embodiment of the present invention completely initializes the kernels by using values of -1 to 1 when initially training using the convolutional neural network 510, and data having a negative value In the case of using an activation function that outputs all 0s, output values including valid information cannot be transferred to the next convolution layer, resulting in a decrease in learning efficiency. Accordingly, in an embodiment of the present invention, the learning unit 331 adds a Leacky ReLU layer after the convolution layer A1, and outputs a positive value as it is, but a constant slope for negative input data Can be output to have. Through this, the learning unit 331 can prevent a problem of a decrease in convergence speed and a local minimization during learning.

The input 2501 may be a set of input feature maps to which padding has been applied, and padding refers to a technique of filling a partial area of the input with a specific value. Specifically, applying padding to the input with the size of the pad as 1 means filling a specific value on the edge of the input feature map, and zoro padding means setting the specific value to 0. it means. For example, if zero padding with a pad size of 1 is applied to the size input of X x Y x Z, the padded input has an edge of 0 and a size of (X+1) x As data of (Y+1) x Z, it may include (X+1) x (Y+1) x Z input elements.

Meanwhile, the learning unit 331 learns by using a time correlation of a plurality of images in performing a convolution operation by inputting a plurality of images including speckle information. In this case, each of the images It may include a plurality of speckles that are grain-shaped patterns. In this case, the learning unit 331 may learn the classification criteria based on the temporal correlation of each speckle. In other words, the learning unit 331 includes information of time, not two-dimensional information of an image. Classification criteria are learned using three-dimensional information.

The learning unit 331 should focus on information on one speckle, and in order to accurately acquire three-dimensional information on one speckle, the learning unit 331 classifies the one speckle and the surrounding speckles to determine the classification criteria. You have to learn. Accordingly, the learning unit 331 may perform a convolution operation using a convolution kernel having a size smaller than the size of one speckle. That is, when the size of one speckle corresponds to m pixels, the learning unit 331 performs a convolution operation using a convolution kernel having an n x n size smaller than m. As an embodiment, as described above, since the image sensor 300 is arranged such that a maximum of 5 or less pixels are located in a speckle grain size, m may be 5, where n is 1 Can have a value. That is, the learning unit 331 may perform a convolution operation using a 1 x 1 convolution kernel. However, the technical idea of the present invention is to perform a convolution operation using a kernel having a size smaller than the size of the speckle, but is not limited thereto.

The kernel set S600 includes convolution kernels corresponding to D output channels, and each convolution kernel may include kernel feature maps corresponding to the number of images. Since the size of each kernel feature map is n x n, the kernel set S600 includes n x n x C x D kernel elements. Here, the size of the convolution kernel is n x n x C, and C may be the number of a plurality of images, that is, the number of image frames. The number of convolution kernels may be the same as the number C of a plurality of images, and in this case, the output result of each layer may be used for Fourier transform under the same condition or analysis using the same.

As shown in FIG. 9, the learning unit 331 performs an operation between the input 2501 and the convolution kernel corresponding to the first output channel of the kernel set S600 to perform an output feature map corresponding to the first output channel ( 25111) can be created. In this way, the learning unit 331 may generate output feature maps 25111 corresponding to the D output channels by performing an operation between the D kernels and the input 2501 in the kernel set S600, respectively. And, it is possible to generate the output (22511) including the generated output feature map (25111).

For example, the learning unit 331 performs an operation between the convolution kernel 600 corresponding to the D-th output channel having a size of nxnx C and an input 501 having a size of W x H x C to obtain W x H An output feature map 25111 having the size of may be generated, and the generated output feature map 5111 corresponds to a D-th output channel. Specifically, the convolution kernel 600 corresponding to the D-th output channel includes C kernel feature maps, and the size of each kernel feature map is n x n. The learning unit 331 slides each kernel feature map having a size of nxn on each input feature map having a size of W x H included in the input 501 to a specific stride, and corresponds to the D-th output channel. An output feature map 25111 which is a result of an operation between the convolution kernel 2600 and the input 2501 may be generated.

Here, the stride means an interval for sliding the kernel feature map during a convolution operation. As described above, since the learning unit 331 needs to learn the classification criteria by classifying the one speckle and the surrounding speckles in order to focus on information on one speckle, the sliding interval stride (s ) May have a value corresponding to the size of the convolution kernel so that each convolution kernel and a region corresponding thereto do not overlap. In other words, when a convolution kernel of size n x n is used, stride (s) may have a value of n. For example, in the case of a 1 x 1 convolution kernel, stride (s) may be 1. Through this, the learning unit 331 may allow one speckle to be non-overlapping with other speckles around it, so that the classification criterion may be learned using time information of only one speckle.

As an embodiment, in learning the classification criteria using the convolutional neural network 510, the learning unit 331 performs a convolution operation to have the same number of output channels (D) as the number of images (C). I can. In other words, the kernel set S600 may include D convolution kernels including C kernel feature maps, and in this case, C and D may be the same.

Meanwhile, the learning unit 331 may reduce the size of the output 2515 by using a pooling operation after a convolution operation using the convolution layers A1. For example, the learning unit 331 may reduce the size of the output 2515 by using sub sampling (A2). For example, sub-sampling may be global average pooling, which is an operation that sets an average value within a range of a certain size as a representative of the range. However, the present invention is not limited thereto, and of course, maximum pooling, min pooing, and the like may be used.

Thereafter, the learning unit 331 applies a weight with a predetermined operation to the feature maps 521 extracted by passing the sub-sampling (A2) filter using a fully connected layer, and then outputs the final output. (2541) can be obtained. For example, the learning unit 331 further applies Leaky ReLU to a fully connected layer after subsampling A2 (A2) (A3), and then, a fully connected layer (Fully connected layer). layer) by applying a softmax layer (A4) to obtain a final output. Here, the final output 2451 may be a microorganism classification criterion for classifying the type or concentration of microorganisms according to the time correlation of the speckle.

Thereafter, the server 30 may receive an image about the wave speckle obtained by irradiating the wave on the new collected liquid W from the airborne bacteria measuring device 10, and the determination unit 333 The type or concentration of suspended bacteria contained in the new collection liquid (W) can be classified based on one microorganism classification standard. However, the determination unit 333 does not necessarily need to be included in the server 30, and when the microorganism classification standard learned from the server 30 is provided to the control unit 400 of the airborne bacteria measuring device 10, the control unit ( 400), like the function of the determination unit 333, may classify the type or concentration of suspended bacteria included in the new collection liquid W based on the microorganism classification criteria.

13 is a graph comparing predicted microbial information (prediction) obtained by measuring airborne bacteria through machine learning according to an embodiment of the present invention and actual microbial information (ground truth).

Referring to FIG. 13, it can be seen that the matching rate between the expected microbial information obtained through the airborne microbial measurement technology according to an embodiment of the present invention and the actual microbial information is very high. 13, the airborne bacteria measurement device 10 uses the microbial classification criteria to determine the types of microorganisms contained in the collection liquid W (B.subtilis, E.Coli, P.aeruginosa, S.aureus) ), etc., as well as each concentration.

As described above, the apparatus for measuring airborne bacteria according to embodiments of the present invention collects airborne bacteria in the collection liquid, and then detects the presence of microorganisms in the collection liquid using a wave speckle. Airborne bacteria can be quickly and accurately detected without chemical methods. In addition, the airborne bacteria measurement device obtains a classification standard for classifying the type or concentration of microorganisms using the change in the time correlation of the speckle through machine learning, thereby quickly and accurately determining the type or concentration of airborne bacteria. Can be distinguished. Through this, it is possible to determine the presence or absence of pathogens in the air in the region, and to effectively check regional quarantine information through the network environment.

So far, we have looked at the center of the preferred embodiment for the present invention. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered from an illustrative point of view rather than a limiting point of view. The scope of the present invention is shown in the claims rather than the above description, and all differences within the scope equivalent thereto should be interpreted as being included in the present invention.

10: airborne bacteria measurement device
100: wave source
300: image sensor
400: control unit
1000: collection unit

Claims (1)

A collection having a storage tank for receiving the collection liquid therein, an intake flow path for sucking external air to one side of the storage tank to guide the collected liquid, and an exhaust flow channel for discharging the air of the storage tank to the outside on the other side of the storage tank unit;
A wave source for irradiating a wave toward the collecting liquid of the collecting unit;
An image sensor that measures a wave speckle generated by multiple scattering in the collection liquid in time series; And
Airborne bacteria measurement apparatus comprising a; a control unit for detecting the presence or absence of microorganisms in the collection liquid based on the change over time of the measured wave speckle.

Images