KR20190093906A - Apparatus for testing generalized likelihood ratio in Fourier-transform infrared spectroscopy, Method thereof, and Computer readable storage medium having the method - Google Patents

Abstract

The present invention relates to an infrared spectroscopy spectrum technology and, more specifically, to a generalized likelihood ratio testing apparatus and method for detecting Fourier-transform infrared (FT-IR) based remote chemical gas. According to the present invention, statistical analysis of various backgrounds and interference material spectrum and gas-specific standard spectrum library data is performed rather than analyzing a single measured spectrum to find the chemical gas, and gas contaminant plume can be detected by reversely tracking a parameter (target gas) having the highest likelihood with respect to the measured spectrum.

Description

Apparatus for testing generalized likelihood ratio in Fourier-transform infrared spectroscopy, Method approx, and computer readable storage medium having the method}

The present invention relates to infrared spectroscopic spectral technology, and more particularly, to a general likelihood ratio assay device and method for Fourier-transform infrared (FT-IR) based far chemical gas detection.

In addition, the present invention relates to a general likelihood ratio assay apparatus and method for detecting a gas polluted cloud using a probability likelihood Maximum Likelihood Method using target gas spectrum and non-target gas spectrum data.

In the case of the conventional Fourier-transform infrared (FT-IR) based long range gas detection, the similarity between the measured spectrum and the standard library was determined to determine whether to detect the chemical gas. This conventional technique can easily cause distortion from signal noise caused by detection equipment (spectral) and interference signals from various backgrounds (steam, infrared, soot, smoke, etc.).

In addition, this distortion causes a problem that the detection sensitivity and / or reliability is greatly reduced. Accordingly, there is a need for a surveillance detection technique that can improve detection sensitivity and / or reliability.

1. Korea Registered Patent No. 10-1469071 (2014.11.28) 2. Korea Patent Registration No. 10-0807441 (2008.02.19) 3. Japanese Patent No. 5729332 (2015.04.17)

1. Cho, Nam, et al., "Quantitative Analysis of Remote Fire Combustion Gases Using Passive Open-Path FT-IR Spectrometer," Korean Society of Analytical Science, Vol. 2. Yu-Jin Jung et al., "Real-time Detection of SF6 Pollution Clouds Using Passive FTIR Remote Chemistry Detector," Journal of the Korean Institute of Military Science and Technology, Vol. 17, No. 1, pages 8-14, 2014

The present invention has been proposed to solve the problems according to the above background, and an object of the present invention is to provide a general likelihood ratio assay device and method that enables detection of far-transform infrared (FT-IR) -based far-field chemical gases.

In addition, the present invention is a general likelihood ratio assay device and method that can detect the gas contaminant plume using the probability likelihood Maximum Likelihood Method using the target gas spectrum and non-target gas spectrum data There is another purpose to provide.

In order to achieve the above-mentioned problem, the present invention is not a method of finding a chemical gas by analyzing a single measured spectrum, but statistically analyzing various background and interference spectra and gas-specific standard spectrum library data. A general likelihood ratio assay device is provided that tracks the highest likelihood parameters (target gases) in reverse.

The general likelihood ratio assay device,

 An acquisition module for obtaining a spectral signal measured from target gases via a spectral sensor;

 A signal processing module for processing the spectral signal into spectral data; And

And an algorithm performing module that extracts a tracking like gas having the highest likelihood among the target gases by performing a predetermined general likelihood ratio test (GLRT) algorithm on the spectral data. do.

In addition, the general likelihood ratio assay device may further include a detection module for comparing the tracking target gas with a predetermined specific value and determining the tracking target gas as a gas pollution cloud according to a comparison result. .

The specific value may be set for each gas through indoor and outdoor tests, and may be a fluid value for each target gas.

In addition, the spectral data may be composed of Fourier-transform infrared (FT-IR) background and interference spectrum data and target gas spectrum data.

In addition, the general likelihood ratio test (GLRT) algorithm uses a general likelihood ratio test matrix, wherein the general likelihood ratio test matrix is the target gases arranged in the first column and the background and interference arranged in rows and columns corresponding to the target gases. The sub-matrix is made of materials.

In addition, the sub-matrix may be a combination of a background spectrum measured in various environments and a spectrum of interference materials having a signal characteristic similar to a target gas to be detected.

In addition, the target gases arranged in the first column may be characterized by being an absorption coefficient vector.

The apparatus may further include a photographing imager photographing the shape of the gas pollution cloud.

In addition, the general likelihood ratio assay apparatus may further include a display for outputting the shape of the gas pollution cloud.

On the other hand, another embodiment of the present invention comprises the steps of: (a) acquiring the spectral signal from the target gases by the acquisition module; (b) a signal processing module signal processing the spectral signal into spectral data; And (c) the algorithm performing module performing a predetermined generalized likelihood ratio test (GLRT) algorithm on the spectral data to extract the highest likelihood tracking target gas among the target gases; We can provide a general likelihood ratio test method in Fourier-transform infrared (FT-IR) based infrared spectroscopy spectrum.

In addition, in the general likelihood ratio assay method, after the step (c), (d) the detection module 440 compares the tracking target gas with a predetermined specific value k and according to a comparison result according to the comparison target gas. Determining as a gas pollution cloud may be characterized in that it further comprises.

In addition, the step (d) may further comprise the step of photographing the image capture the shape of the gas pollution cloud.

The step (d) may further include outputting a shape of the gas polluted cloud on a display.

On the other hand, another embodiment of the present invention can provide a computer readable storage medium storing program code for executing the general likelihood ratio test method in the FT-IR based infrared spectroscopy spectrum described above. .

 According to the present invention, rather than analyzing a single measured spectrum to find chemical gases, statistically analyzing various background and interference spectra and gas-specific standard spectral library data, the most likelihood of the measured spectrum is obtained. By tracking back high parameters (target gases), gas cloud clouds can be detected.

In addition, another effect of the present invention is to analyze the interference factors statistically through the Generalized Likelihood Ratio Test (GLRT) to determine the highest likelihood (target gas) for the measured spectrum. By tracking, the performance degradation caused by various interference factors can be minimized.

1 is a conceptual diagram of a general likelihood ratio test matrix according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a gas detection process using a generalized likelihood ratio test (GLRT) algorithm in a Fourier-transform infrared (FT-IR) based spectral spectrum according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating a process of a general likelihood ratio test algorithm used in the general likelihood ratio test algorithm performing step S230 shown in FIG. 2 in detail.
FIG. 4 is a block diagram illustrating a general likelihood ratio assay device in Fourier-transform infrared (FT-IR) based spectral spectra according to an embodiment of the present invention.
5 is an experimental result graph showing the effect of improving the detection algorithm according to an embodiment of the present invention.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention.

In describing each drawing, like reference numerals are used for like elements.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term “and / or” includes any combination of a plurality of related items or any item of a plurality of related items.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art.

Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and are not construed in ideal or excessively formal meanings unless expressly defined in this application. Should not.

Hereinafter, a general likelihood ratio assay device and method in Fourier-transform infrared (FT-IR) based infrared spectroscopy spectrum according to an embodiment of the present invention will be described in detail.

…

으로 표현되며, Z는 S₁…S_n,

…

으로 표현된다. BG는 배경 및 간섭물질을 나타내고, S는 표적가스를 나타낸다. 일반적으로 배경

행렬(matrix)에는 초목, 하늘, 바다 등 다양한 환경에서 측정한 배경 스펙트럼과 탐지하고자 하는 표적가스와 유사한 신호 특성을 보이는 간섭물질들의 스펙트럼 조합으로 구성되어진다. 여기에 탐지하고자 하는 표적가스의 흡수계수(absorption coefficients) 벡터(vector)를 더하게 되면 일반 우도비 검정 행렬인

행렬(matrix)을 구할 수 있다. 이러한 Z행렬의 예를 보여주는 도면이 도 1에 도시된다. 1 is a conceptual diagram of a general likelihood ratio test matrix according to an embodiment of the present invention. Referring to Figure 1, the B matrix is

…

Where Z is S ₁ . S _n ,

…

It is expressed as BG represents background and interferences, and S represents target gas. Background in general

The matrix consists of a combination of background spectra measured in a variety of environments, including vegetation, sky and sea, as well as spectra of interfering substances with signal characteristics similar to those of the target gas to be detected. Adding the vector of the absorption coefficients of the target gas to be detected, the normal likelihood ratio test matrix

You can get a matrix. A diagram showing an example of such a Z matrix is shown in FIG.

FIG. 2 is a flowchart illustrating a gas detection process using a generalized likelihood ratio test (GLRT) algorithm in a Fourier-transform infrared (FT-IR) based spectral spectrum according to an embodiment of the present invention. Referring to FIG. 2, first, a spectral signal is obtained using a spectrum sensor (step S210).

Thereafter, the spectral signal is subjected to a signal preprocessing step such as signal correction (step S220).

Thereafter, the signal-processed spectrum data is performed using a generalized likelihood ratio test (GLRT) algorithm (step S230, Equation 6).

Then, when the result of the algorithm (y) is higher than the specific threshold value k, the corresponding gas is detected (steps S240, S250, Equation 7).

행렬와

행렬(도 1) 구축 과정을 상세하게 보여주는 흐름도이다. 도 3을 참조하면, 일반 우도비 검정기법(GLRT : Generalized Likelihood Ratio Test)은 통계적 가설 검정 방법을 기본으로 하고 있다. 부연하면, 일반 우도비 검정기법은 FT-IR(Fourier-transform infrared) 배경 및 간섭물질 스펙트럼, 표적가스 스펙트럼 데이터를 근거로 하여 모집단에 대한 가설이 맞는지를 통계적으로 검정하는 분석기법의 일종이다(단계 S310,S320). 3 is a background used in the general likelihood ratio test algorithm performing step (S230) shown in FIG.

Matrix and

1 is a flowchart showing in detail the process of constructing a matrix (FIG. 1). Referring to FIG. 3, the generalized likelihood ratio test (GLRT) is based on the statistical hypothesis test method. In other words, the general likelihood ratio test is a type of analysis method that statistically tests the hypothesis of a population based on Fourier-transform infrared (FT-IR) background, interference spectra, and target gas spectral data. S310, S320).

The measured FT-IR spectrum can be represented by two hypotheses (H ₀ , H ₁ ). This is expressed as the following equation.

는 기존의 FT-IR 온도 스펙트럼으로 구성된 행렬(matrix)형태의 배경 및 간섭물질 데이터이다.

는 탐지하고자하는 표적가스의 스펙트럼이다.

는 가우시안(Gaussian) 잡음(noise)이며, 잡음 자체는 확률 변수(random variable) 이므로 측정된 신호(

) 자체도 확률적인 성질을 갖는다. 따라서 정규분포(Normal distribution) 또는 가우시안분포(Gaussian distribution)로 평균이 0이고 표준편차가 σ인 N(0,σ²I)로 표현할 수 있다. 여기서 I는 단위행렬이다.

는 알려지지 않은 상수 벡터(unknown vector)의 성격을 갖는다. 따라서 측정된 FT-IR(Fourier-transform infrared) 스펙트럼

는 다음식과 같은 가우시안(Gaussian) 분포를 갖는다.here,

Is background and interference data in a matrix form of the existing FT-IR temperature spectrum.

Is the spectrum of the target gas to be detected.

Is Gaussian noise, and the noise itself is a random variable, so the measured signal (

) Itself is also stochastic. Therefore, it can be expressed as N (0, σ ² I) having a mean of 0 and a standard deviation of σ in a normal distribution or a Gaussian distribution. Where I is the unit matrix.

Has the nature of an unknown vector. Fourier-transform infrared (FT-IR) spectra thus measured

Has a Gaussian distribution as

Here, B is a matrix composed of a background and an interference spectrum, and Z is a matrix in which B and the target gas spectrum are combined.

에 대한 우도(likelihood) 확률(P)은 다음식과 같이 구할 수 있다.Also,

The likelihood probability (P) for can be obtained as

Where T stands for transpose.

Likelihood test method using the log function can be obtained from the above equation. This is expressed as the following equation.

를 다음식과 같이 구할 수 있다. To maximize the difference in probability

Can be obtained as

를 우도비 검정기법(LRT : Likelihood Ratio Test)에(수학식4) 대입하게 되면 수학식6와 같은 등가의 검정 통계 기법(D)을 구할 수 있다. So obtained

By substituting the likelihood ratio test (LRT) (Equation 4), an equivalent test statistical technique (D) can be obtained.

는 각 배경 스펙트럼과 탐지하고자 하는 가스 스펙트럼에 대한 직교(orthogonal) 성질을 가진 행렬(matrix)이다. 즉, 센서 장비로부터 측정한 스펙트럼(

)와 미리 구성한

행렬(matrix)를 활용하여 수학식6에 따라 상수 결과 값(

)을 획득할 수 있다. 여기서 구한 결과 값은 특정 임계치 값(

) 과 비교하여 다음식과 같은 최종적인 판별을 수행한다. here

Is a matrix with orthogonal properties for each background spectrum and gas spectrum to be detected. That is, the spectrum measured from the sensor equipment (

) And preconfigured

Using a matrix, the constant result value according to equation (6)

) Can be obtained. The resulting value is a specific threshold value (

) And final decision is made as follows.

가 k보다 크거나 같으면 표적 가스가 존재하는 것이고,

가 k보다 작으면 표적 가스가 존재하지 않는 것으로 판별된다.In the above equation

Is greater than or equal to k, the target gas is present,

Is less than k, it is determined that no target gas is present.

)은 정해진 값이 아닌 탐지하고자 하는 표적가스마다 유동적인 값을 가지므로, 다양한 실내외 시험을 통하여 가스별로 설정이 필요하다(단계 S330, S340).In other words, a specific threshold value (

) Has a fluid value for each target gas to be detected, not a predetermined value, and therefore, setting is required for each gas through various indoor and outdoor tests (steps S330 and S340).

행렬(matrix)의 구성요소가 중요하다.

행렬(matrix)의 구성방법은 도 3에 명시되어 있으며, 어떻게 구성하느냐에 따라 수학식6의 결과 값과 수학식7의 임계값이 달라진다. 이에, 탐지하고자하는 가스마다 각각의

행렬(matrix) 구성이 필요하다. In order to increase the efficiency of the common likelihood ratio test technique for Fourier-transform infrared (FT-IR) gas detection,

The components of the matrix are important.

The method of constructing the matrix is shown in FIG. 3, and the resultant value of Equation 6 and the threshold value of Equation 7 vary depending on how the matrix is constructed. Therefore, each gas to be detected

Matrix construction is required.

FIG. 4 is a block diagram illustrating a general likelihood ratio assay device 400 in Fourier-transform infrared (FT-IR) based spectral spectrum according to an embodiment of the present invention. Referring to FIG. 4, the general likelihood ratio verification apparatus 400 may include an acquisition module 410 which acquires a spectral signal measured from target gases through a spectral sensor 40, and signal processing the spectral signal into spectral data. A signal processing module 420, an algorithm performing module for extracting the highest likelihood tracking target gas among the target gases by performing a preset general likelihood ratio test (GLRT) algorithm on the spectral data 430, and a detection module 440 for comparing the tracking target gas with a predetermined specific value k and determining the tracking target gas as a gas pollution cloud according to a comparison result.

The spectrum sensor 40 detects and measures the infrared radiation signal of the natural background according to the gas pollution cloud existing in the atmosphere. The infrared light received from the natural background and the target (target) gas is transmitted to an interferometer (not shown) through an optical system (not shown) inside the scanning device, and is then detected by a light splitter (not shown) and a high speed rotating plate inside the interferometer. C) generates an interferogram.

The acquisition module 410 collects the interference signal (that is, the spectrum signal) generated by the spectrum sensor 40.

The signal processing module 420 first amplifies and filters the interference fringes generated by the spectral sensor 40 (ie, spectral signals) through analog-to-digital (AD) conversion, is stored as a digital signal, and then peaked. After the detection, the spectral spectrum of the wavenumber region is obtained by Fourier transform. Of course, AD conversion can be performed using an AD converter and can be performed using software.

The algorithm execution module 430 performs a function of performing a generalized likelihood ratio test (GLRT) algorithm generated by FIG. 3.

The detection module 440 determines a target gas extracted through a generalized likelihood ratio test (GLRT) algorithm as a pollution cloud and outputs it. Of course, the display 450 may be connected to the detection module 440 for this purpose.

The display 450 may be a liquid crystal display (LCD), a light emitting diode (LED) display, a plasma display panel (PDP), an organic LED (OLED) display, a touch screen, a cathode ray tube (CRT), a flexible display, or the like. have.

In another embodiment of the present invention, it is also possible to output an image of the target gas through a display in conjunction with an imager (not shown). The imager may be a closed circuit television (CCTV), a charge-coupled device (CCD) camera, a complementary metal-oxide semiconductor (CMOS) camera, or the like.

The term "... module" described in the specification means a unit for processing at least one function or operation, which may be implemented in hardware and / or software. In hardware implementation, an application specific integrated circuit (ASIC), a digital signal processing (DSP), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, and a microprocessor are designed to perform the above functions. , Other electronic units, or a combination thereof. In the software implementation, the module may be implemented as a module that performs the above-described function. The software may be stored in a memory unit and executed by a processor. The memory unit or processor may employ various means well known to those skilled in the art.

5 is an experimental result graph showing the effect of improving the detection algorithm according to an embodiment of the present invention. Referring to FIG. 5, it is a comparison between the conventional signal similarity technique (Cross Correlation) 510 and the sample gas low concentration gas detection probability of the general likelihood ratio check method 520 according to an embodiment of the present invention. In general, it is difficult to detect gas at low concentrations. Samples 1 and 2 used in the test are SF ₆ and Freon, which are simulation agents. Of course, other agents are also available.

In the case of the general likelihood ratio inspection technique 520, the detection probability of the sample gas 1 and the sample gas 2 exceeds 0.95. In contrast, in the case of the signal similarity technique 510, the detection probability of sample gas 1 is only about 0.8, and the detection probability of sample gas 2 is slightly higher than 0.85.

In addition, the steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in a program instruction form that can be executed by various computer means and recorded in a computer-readable medium. The computer readable medium may include program (instruction) code, data file, data structure, etc. alone or in combination.

The program (command) code recorded on the medium may be those specially designed and configured for the present invention, or may be known and available to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, Blu-rays, etc., and ROMs and RAMs. Semiconductor memory devices specifically configured to store and execute program (command) code, such as RAM), flash memory, and the like.

Here, examples of program (instruction) code include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

40: spectrum sensor
400: general likelihood ratio device
410: acquisition module
420: signal processing module
430: algorithm execution module
440: detection module
450: display

Claims (14)

An acquisition module 410 for acquiring the spectral signal measured from the target gases via the spectral sensor 40;
A signal processing module 420 for signal processing the spectral signal into spectral data; And
An algorithm performing module 430 for extracting a tracking like gas having the highest likelihood among the target gases by performing a predetermined generalized likelihood ratio test (GLRT) algorithm on the spectral data;
A common likelihood ratio assay device in Fourier-transform infrared (FT-IR) based infrared spectral spectra comprising a.
The method of claim 1,
And a detection module 440 for comparing the tracking target gas with a preset specific value k and determining the tracking target gas as a gas pollution cloud according to a comparison result. General likelihood ratio assay device in spectral spectrum.
The method of claim 2,
The specific value (k) is set for each gas through the indoor and outdoor tests, the general likelihood ratio assay device in the FT-IR-based infrared spectroscopy, characterized in that the value is fluid for each target gas.
The method of claim 1,
And the spectral data comprises a Fourier-transform infrared (FT-IR) background and interference spectrum data and target gas spectral data.
The method of claim 3, wherein
The general likelihood ratio test (GLRT) algorithm uses a general likelihood ratio test matrix (Z), wherein the general likelihood ratio test matrix is a target gas arranged in a first column and a background arranged in rows and columns corresponding to the target gas and A general likelihood ratio assay device in FT-IR based infrared spectroscopy characterized by a submatrix (B) of interferences.
The method of claim 5,
The submatrix (B) is a common likelihood ratio test in FT-IR based infrared spectroscopy, characterized in that it is a combination of a background spectrum measured in various environments and a spectrum of interfering substances having a signal characteristic similar to a target gas to be detected. Device.
The method of claim 5,
The likelihood ratio assay device in an FT-IR based infrared spectral spectrum, wherein the target gases arranged in the first column are absorption coefficients vectors.
The method of claim 2,
And a photographing imager for photographing the shape of the gas pollution cloud.
The method of claim 8,
And a display for outputting the shape of the gas polluted cloud. The likelihood ratio assay device of FT-IR-based infrared spectroscopy. (a) the obtaining module 410 obtaining the spectral signal measured from the target gases via the spectral sensor 40;
(b) the signal processing module 420 signal processing the spectral signal into spectral data; And
(c) an algorithm performing module 430 extracts a tracking like gas having the highest likelihood among the target gases by performing a predetermined likelihood ratio test (GLRT) algorithm for the spectral data. step;
General likelihood ratio assay method in Fourier-transform infrared (FT-IR) based infrared spectral spectra.
The method of claim 10,
After the step (c), (d) the detection module 440 compares the tracking target gas with a predetermined specific value (k) and determines the tracking target gas as a gas pollution cloud according to a comparison result; A general likelihood ratio assay method in FT-IR based infrared spectroscopy, further comprising.
The method of claim 11,
In the step (d), the photographing imager further comprises the step of capturing the shape of the gas pollution cloud; general likelihood ratio test method in the FT-IR-based infrared spectral spectrum.
The method of claim 12,
The step (d), the likelihood ratio calibration method of the FT-IR-based infrared spectral spectrum, characterized in that it further comprises the step of outputting the shape of the gas pollution cloud on the display.
A computer-readable storage medium storing program code for executing a general likelihood ratio assay method on an FT-IR based infrared spectral spectrum according to any one of claims 10 to 13.

Images