[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

JP2006126100A - Discrimination method using near-infrared spectrum - Google Patents

Discrimination method using near-infrared spectrum Download PDF

Info

Publication number
JP2006126100A
JP2006126100A JP2004317434A JP2004317434A JP2006126100A JP 2006126100 A JP2006126100 A JP 2006126100A JP 2004317434 A JP2004317434 A JP 2004317434A JP 2004317434 A JP2004317434 A JP 2004317434A JP 2006126100 A JP2006126100 A JP 2006126100A
Authority
JP
Japan
Prior art keywords
discrimination
infrared spectrum
line
abnormal
normal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2004317434A
Other languages
Japanese (ja)
Inventor
Naoyoshi Morita
尚喜 森田
Tatsuya Kojima
達也 小島
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kao Corp
Original Assignee
Kao Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kao Corp filed Critical Kao Corp
Priority to JP2004317434A priority Critical patent/JP2006126100A/en
Publication of JP2006126100A publication Critical patent/JP2006126100A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of drawing a discrimination line drawable even in the case of no abnormal article, and enabling everybody to discriminate between an abnormal article and a normal article simply and accurately, and a discrimination method using the discrimination line. <P>SOLUTION: The method of drawing the discrimination line for discriminating an abnormal article includes [1] (a) a process for measuring a near-infrared spectrum of the normal article, (b) a process for calculating a standard spectrum and a standard deviation (σ) based on the near-infrared spectrum acquired in the process (a), and (c) a process for drawing the discrimination line based on the standard deviation (σ) acquired in the process (b). This discrimination method of the abnormal article includes (d) a process for measuring the near-infrared spectrum of a discrimination object sample, and (e) a process for comparing the absolute value of a deviation of the near-infrared spectrum acquired in the process (d) to the standard spectrum acquired in the process (b) of [1] with the discrimination line drawn by the method described in [1]. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

本発明は、近赤外スペクトルを用いて異常品を判別するための判別ラインの作成方法および当該判別ラインを用いた異常品の判別方法に関する。   The present invention relates to a method for creating a discrimination line for discriminating an abnormal product using a near-infrared spectrum and a method for discriminating an abnormal product using the discrimination line.

近年、近赤外分光法による評価方法は非破壊測定方法として注目され、過去20年間に数多くの研究がなされ、実用化が進められている。例えば、食品業界では正常肉と異常肉との判別に近赤外分光法を利用する方法が報告されている(例えば、特許文献1参照)。   In recent years, the evaluation method by near infrared spectroscopy has attracted attention as a nondestructive measurement method, and many studies have been made in the past 20 years, and its practical application has been promoted. For example, in the food industry, a method using near infrared spectroscopy for discrimination between normal meat and abnormal meat has been reported (for example, see Patent Document 1).

また、製造業において、原料の異常が原因で、異常品ができてしまうことが少なくない。製造業においては製品化することによりコストが大きくなるため、製品化する前に原料の異常を検知することは重要である。しかし、受入検査項目をクリアした原料でも異常品ができてしまうことがあるため、原料の異常を検知する方法が求められていた。
特開2002−328088号公報
In the manufacturing industry, abnormal products are often produced due to abnormalities in raw materials. In the manufacturing industry, since the cost is increased by commercialization, it is important to detect abnormalities in raw materials before commercialization. However, even if the raw material has cleared the acceptance inspection items, an abnormal product may be produced. Therefore, a method for detecting the abnormality of the raw material has been demanded.
JP 2002-328088 A

しかしながら、近赤外分光法で一般的に用いられる判別分析の手法では、必ずその品種の正常品だけでなく異常品を用いて判別モデルを作成する必要があるため、この方法を、多くの品種を用いる産業において判別のために用いるためには、全ての品種の異常品を準備する必要があり、事実上不可能である。   However, discriminant analysis methods generally used in near-infrared spectroscopy must always create discriminant models using not only normal products but also abnormal products. In order to use for discrimination in the industry that uses, it is necessary to prepare abnormal products of all varieties, which is practically impossible.

また、仮に異常品を準備することができて判別モデルを作成したとしても、別の原因による異常品が出現した場合には、従来の方法では検知できないという問題があった。   Further, even if an abnormal product can be prepared and a discrimination model is created, if an abnormal product due to another cause appears, there is a problem that it cannot be detected by the conventional method.

さらに、正常品と異常品を用いて判別モデルが作成されたが、正常品と異常品との差が小さい場合は、誰にでも正確に判別できるわけではないという問題もあった。   Furthermore, a discrimination model is created using normal products and abnormal products. However, when the difference between normal products and abnormal products is small, there is a problem that no one can accurately discriminate.

従って、本発明は、異常品がない場合でも作成でき、簡便にかつ正確に、誰にでも異常品と正常品とを判別できる判別ラインの作成方法および当該判別ラインを用いた判別方法を提供することを目的とする。   Therefore, the present invention provides a determination line creation method and a determination method using the determination line, which can be created even when there is no abnormal product, and can be easily and accurately discriminated between an abnormal product and a normal product by anyone. For the purpose.

すなわち、本発明は、
〔1〕a)正常品の近赤外スペクトルを測定する工程、
b)工程a)で得られた近赤外スペクトルをもとに標準スペクトルおよび標準偏差(σ)を算出する工程、ならびに
c)工程b)で得られた標準偏差(σ)をもとに判別ラインを作成する工程
を含む、異常品を判別するための判別ラインの作成方法、ならびに
〔2〕d)判別対象サンプルの近赤外スペクトルを測定する工程、および
e)前記〔1〕の工程b)で得られた標準スペクトルに対する工程d)で得られた近赤外スペクトルとの偏差の絶対値と、前記〔1〕記載の方法により作成された判別ラインとを比較する工程
を含む、異常品の判別方法
に関する。
That is, the present invention
[1] a) a step of measuring a near-infrared spectrum of a normal product,
b) A step of calculating a standard spectrum and a standard deviation (σ) based on the near-infrared spectrum obtained in step a), and c) A discrimination based on the standard deviation (σ) obtained in step b). A method of creating a discrimination line for discriminating abnormal products, including a step of creating a line, [2] d) a step of measuring a near-infrared spectrum of a discrimination target sample, and e) a step b of the above [1] The abnormal product including the step of comparing the absolute value of the deviation from the near-infrared spectrum obtained in step d) with respect to the standard spectrum obtained in step 1) and the discrimination line created by the method described in [1] above. It is related with the discriminating method.

本発明により、異常品がない場合でも作成でき、簡便にかつ正確に異常品と正常品とを判別できる判別ラインの作成方法および当該判別ラインを用いた判別方法を提供することができる。   According to the present invention, it is possible to provide a method for creating a discrimination line that can be created even when there is no abnormal product, and that can easily and accurately discriminate between an abnormal product and a normal product, and a discrimination method using the discrimination line.

本発明の異常品を判別するための判別ラインの作成方法は、
a)正常品の近赤外スペクトルを測定する工程、
b)工程a)で得られた近赤外スペクトルをもとに標準スペクトルおよび標準偏差(σ)を算出する工程、ならびに
c)工程b)で得られた標準偏差(σ)をもとに判別ラインを作成する工程
を含むことに一つの大きな特徴を有する。
A method for creating a discrimination line for discriminating abnormal products according to the present invention is as follows.
a) a step of measuring a near-infrared spectrum of a normal product,
b) A step of calculating a standard spectrum and a standard deviation (σ) based on the near-infrared spectrum obtained in step a), and c) A discrimination based on the standard deviation (σ) obtained in step b). One major feature is that it includes the step of creating a line.

かかる特徴を有することにより、従来のように正常品および異常品を用いて判別モデルを作成した後でなければサンプルを判別できないといった欠点が解消され、品質検査試験、原料の受入れ試験にかかる手間が顕著に減少され、種々の原因による異常品を判別することができる。   By having such characteristics, the disadvantage that samples cannot be discriminated only after creating a discrimination model using normal and abnormal products as in the past is eliminated, and the labor involved in quality inspection tests and raw material acceptance tests is eliminated. Remarkably reduced, abnormal products due to various causes can be identified.

本発明において「判別ライン」とは、正常品と異常品との閾値のことをいう。   In the present invention, the “discrimination line” refers to a threshold value between a normal product and an abnormal product.

本発明において「正常品」とは判別ラインを用いて判断した場合に、判別ライン以内のものをいい、「異常品」とは正常品以外の全てのものをいう。   In the present invention, “normal product” refers to items within the determination line when determined using the determination line, and “abnormal product” refers to all items other than the normal product.

「判別ライン以内」とは、判別ラインから全波数において逸脱していないことをいう。   “Within the discrimination line” means that there is no deviation from the discrimination line in all wave numbers.

本発明により判別ラインを作成することができる物質としては、近赤外スペクトルが測定できる物質であれば特に限定されないが、各種有機化合物が挙げられる。各種有機化合物としては、各種界面活性剤に用いられる化合物、食油などの油脂類、農作物、調味料などが挙げられる。具体的には、N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミド、ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液、シリコン被覆酸化亜鉛などが挙げられる。   The substance capable of creating a discrimination line according to the present invention is not particularly limited as long as it is a substance capable of measuring a near-infrared spectrum, and various organic compounds are exemplified. Examples of the various organic compounds include compounds used for various surfactants, fats and oils such as edible oil, agricultural products, seasonings and the like. Specifically, N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide, poly (N-propanoylethyleneimine) graft-dimethylsiloxane / γ-aminopropylmethyl Examples thereof include a siloxane copolymer monoethyl sulfate salt solution and silicon-coated zinc oxide.

工程a)において近赤外スペクトルを測定するために用いられる測定装置としては、例えば、フーリエ変換型近赤外分析装置(例えば、ブラン・ルーベ社製;Infra Prover II)、分散型近赤外分析装置などが挙げられる。   As a measuring apparatus used for measuring the near infrared spectrum in the step a), for example, a Fourier transform type near infrared analysis apparatus (for example, manufactured by Blanc-Loube; Infra Prover II), a distributed near infrared analysis Examples thereof include devices.

測定方法としては、例えば、粉体サンプルの場合は反射法、液体サンプルの場合は透過法などが挙げられる。例えば、サンプル測定にフーリエ変換型近赤外分析装置の1つであるブラン・ルーベ社製インフラプルーバーIIを用いる場合、粉体サンプルを測定する場合は固体測定ユニットSPA(Solid Presentation Accessory)を使用して反射法でサンプルの近赤外スペクトルを測定することができ、液体サンプルを測定する場合は液体測定ユニットLPA(Liquid Presentation Accessory)を使用して透過法でサンプルの近赤外スペクトルを測定することができる。   Examples of the measurement method include a reflection method in the case of a powder sample and a transmission method in the case of a liquid sample. For example, when using the Infraprober II manufactured by Blanc-Loube, which is one of the Fourier transform type near-infrared analyzers for sample measurement, the solid measurement unit SPA (Solid Presentation Accessory) is used when measuring powder samples. The near infrared spectrum of the sample can be measured by the reflection method. When measuring a liquid sample, the near infrared spectrum of the sample is measured by the transmission method using the liquid measurement unit LPA (Liquid Presentation Accessory). be able to.

測定に使用される近赤外光の波数は、4000〜12000cm-1の波数が一般に用いられるが、サンプルの種類によって適宜設定すればよい。また、測定における分解能としては、用いられる装置の能力によっても異なるが、好ましくは16cm-1以下、より好ましくは8cm-1以下である。積算回数は、精度を高くする観点から、多い方が好ましいが、例えば、3回以上が好ましく、5回以上がより好ましく、100回以上がさらに好ましい。 The wave number of near-infrared light used for the measurement is generally 4000 to 12000 cm −1 , but may be appropriately set depending on the type of sample. The resolution in measurement varies depending on the ability of the apparatus used, but is preferably 16 cm −1 or less, more preferably 8 cm −1 or less. The number of integrations is preferably larger from the viewpoint of increasing accuracy, but is preferably 3 times or more, more preferably 5 times or more, and still more preferably 100 times or more, for example.

工程a)の測定に必要な正常品のロット数は、判別ラインを正確にするという観点から、好ましくは少なくとも3ロット、より好ましくは少なくとも5ロット、さらに好ましくは少なくとも10ロットである。   The number of normal product lots required for the measurement in step a) is preferably at least 3, more preferably at least 5 and even more preferably at least 10 lots from the viewpoint of making the discrimination line accurate.

工程a)で得られる近赤外スペクトルとしては、前記波数における吸光度スペクトル、正規化スペクトル、一次微分スペクトルまたは二次微分スペクトルが挙げられ、工程b)で使用される近赤外スペクトルは、サンプルによって適宜選択すればよいが、ベースラインの変動の影響を抑えるという観点から、二次微分スペクトルが好ましい。   The near-infrared spectrum obtained in step a) includes an absorbance spectrum, normalized spectrum, first derivative spectrum or second derivative spectrum at the wave number, and the near-infrared spectrum used in step b) depends on the sample. A secondary derivative spectrum is preferable from the viewpoint of suppressing the influence of baseline fluctuations.

本発明において「標準スペクトル」とは、所定のロット数のサンプルをそれぞれ所定の積算回数で測定して得られた近赤外スペクトルの各測定波数での平均値を用いて描かれた近赤外スペクトルをいう。「標準偏差(σ)」とは、各測定波数での標準偏差をいい、かかる標準偏差は当該分野で公知の方法により算出される。   In the present invention, the “standard spectrum” means a near-infrared image drawn using an average value at each measured wave number of a near-infrared spectrum obtained by measuring a predetermined number of lots of samples at a predetermined integration number. The spectrum. “Standard deviation (σ)” refers to the standard deviation at each measured wave number, and the standard deviation is calculated by a method known in the art.

工程c)において、前記工程b)で算出された標準偏差(σ)をもとに判別ラインを作成する方法を説明する。   In step c), a method of creating a discrimination line based on the standard deviation (σ) calculated in step b) will be described.

工程b)で算出された標準偏差(σ)を用いて各波数におけるKσのラインを描き、そのラインを判別ラインとする。   A line of Kσ at each wave number is drawn using the standard deviation (σ) calculated in step b), and the line is used as a discrimination line.

ここで、「K」は、標準偏差に乗する係数を表し、正常品と異常品とを十分判別できるように統計学的に考慮すると、3が好ましい。正常品と異常品とをさらに確実に判別する観点から、Kは、工程a)で測定された正常品の近赤外スペクトルの標準スペクトルに対する偏差の絶対値が全てKσ以下となるような最小の係数(Kmin)以上であることがより好ましい。Kminは、サンプルによって異なるので一概にはいえず、適宜設定されればよく、正常品の近赤外スペクトルが非常に安定している場合は3より小さくなることもあり得る。 Here, “K” represents a coefficient to be multiplied by the standard deviation, and 3 is preferable when statistically considered so that normal products and abnormal products can be sufficiently distinguished. From the standpoint of more reliably discriminating between normal products and abnormal products, K is the minimum such that the absolute values of deviations from the standard spectrum of the near-infrared spectrum of normal products measured in step a) are all equal to or less than Kσ. It is more preferable that the coefficient (K min ) or more. Since K min varies depending on the sample, it cannot be generally specified, and may be set as appropriate. If the near-infrared spectrum of a normal product is very stable, it may be smaller than 3.

以上のように本発明の異常品の判別方法に使用される判別ラインが作成される。   As described above, the discrimination line used in the method for discriminating abnormal products according to the present invention is created.

次に、本発明の異常品の判別方法について説明する。   Next, the method for discriminating abnormal products according to the present invention will be described.

本発明の異常品の判別方法は、
d)判別対象サンプルの近赤外スペクトルを測定する工程、および
e)前記工程b)で得られた標準スペクトルに対する工程d)で得られた近赤外スペクトルの偏差の絶対値と、前記の方法により作成された判別ラインとを比較する工程を含むことに1つの大きな特徴を有する。
The method for discriminating abnormal products according to the present invention includes:
d) a step of measuring a near-infrared spectrum of a sample to be discriminated; and e) an absolute value of a deviation of the near-infrared spectrum obtained in step d) with respect to the standard spectrum obtained in step b), and the method described above. One major feature is that it includes a step of comparing with the discrimination line created by the above.

工程d)は、正常品の代わりに判別対象サンプルを測定すること以外は前記工程a)と同様に行うことができる。   Step d) can be performed in the same manner as step a) except that the determination target sample is measured instead of the normal product.

工程e)における比較方法は、標準スペクトルに対する工程d)で得られた近赤外スペクトルとの偏差の絶対値が、前記で得られた判別ラインから全波数において逸脱していないかを確認し、全く逸脱していない場合は正常品、1つの波数でも判別ラインから逸脱している場合は異常品と判断される。   The comparison method in step e) confirms that the absolute value of the deviation from the near-infrared spectrum obtained in step d) with respect to the standard spectrum does not deviate from the discrimination line obtained above in all wave numbers, When there is no departure from the discriminant product, it is determined that the product is a normal product.

以下の実施例に先立ち、実施例に使用した原料、装置、測定方法および測定条件をまとめて記載する。
(1)原料
N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミド(粉体);
ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液(液体);または
シリコン被覆酸化亜鉛(使用シリコン:メチルハイドロジェンポリシロキサンとメチルポリシロキサン、粉体)
(2)装置
フーリエ変換型近赤外分析装置(ブラン・ルーベ社製;Infra Prover II)
(3)測定方法
粉体試料:固体測定ユニットSPAによる反射法
液体試料:液体測定ユニットLPAによる透過法
(4)測定条件
測定波数:6500〜10000cm-1
分解能 :16cm-1
Prior to the following examples, the raw materials, apparatuses, measurement methods, and measurement conditions used in the examples are collectively described.
(1) Raw material
N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide (powder);
Poly (N-propanoylethyleneimine) graft-dimethylsiloxane / γ-aminopropylmethylsiloxane copolymer monoethyl sulfate sulfate (liquid); or silicon-coated zinc oxide (silicon used: methylhydrogenpolysiloxane and methylpolysiloxane, powder )
(2) Instrument Fourier transform type near-infrared analyzer (Blan Roubaix; Infra Prover II)
(3) Measurement method Powder sample: Reflection method using solid measurement unit SPA Liquid sample: Transmission method using liquid measurement unit LPA (4) Measurement conditions Measurement wave number: 6500-10000 cm -1
Resolution: 16cm -1

実施例1 N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドについての判別ラインの作成
N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドの正常品6ロットの近赤外スペクトルをそれぞれ3回ずつ測定し、得られた二次微分スペクトルの全測定波数における平均値および標準偏差(σ)を算出し、Kσ(K=3)を描いたが、3σを逸脱する正常品が幾らか見られたので、正常品が全く逸脱しなくなるまでKを変動させたところK=3.2以上で判別ラインを逸脱する正常品はなくなった(図1参照)。従って、N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドにおいては、3.2σを判別ライン(図1における太線)とした。
Example 1 Creation of a discrimination line for N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide
N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide normal lots of 6 normal lots were measured 3 times each, and the obtained second derivative spectra The average value and standard deviation (σ) at all measured wave numbers were calculated, and Kσ (K = 3) was drawn. However, some normal products that deviated from 3σ were found, so the normal products did not deviate at all. When K was varied, there were no normal products that deviated from the discrimination line at K = 3.2 or more (see Fig. 1). Therefore, in N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide, 3.2σ was taken as a discrimination line (thick line in FIG. 1).

実施例2 ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液についての判別ラインの作成
ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液の正常品5ロットの近赤外スペクトルをそれぞれ5回ずつ測定し、得られた二次微分スペクトルの全測定波数における平均値および標準偏差(σ)を算出し、Kσ(K=3)を描いたが、3σを逸脱する正常品が幾らか見られたので、正常品が全く逸脱しなくなるまでKを変動させたところK=4.5以上で判別ラインを逸脱する正常品はなくなった(図2参照)。従って、ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液においては、4.5σを判別ライン(図2における太線)とした。
Example 2 Preparation of Discrimination Line for Poly (N-propanoylethyleneimine) Graft-Dimethylsiloxane / γ-Aminopropylmethylsiloxane Copolymer Monoethyl Sulfate Salt Poly (N-propanoylethyleneimine) Graft-dimethylsiloxane / γ- Measure the near-infrared spectra of 5 normal lots of aminopropylmethylsiloxane copolymer monoethyl sulfate solution 5 times each, and calculate the mean value and standard deviation (σ) of all the measured wave numbers of the obtained second derivative spectra. , Kσ (K = 3) was drawn, but there were some normal products that deviated from 3σ, so when K was varied until the normal products did not deviate at all, it deviated from the discrimination line at K = 4.5 or higher. The normal product disappeared (see Fig. 2). Therefore, in the poly (N-propanoylethyleneimine) graft-dimethylsiloxane / γ-aminopropylmethylsiloxane copolymer monoethyl sulfate solution, 4.5σ was taken as the discrimination line (thick line in FIG. 2).

実施例3 シリコン被覆酸化亜鉛についての判別ラインの作成
シリコン被覆酸化亜鉛の正常品3ロットの近赤外スペクトルをそれぞれ12回ずつ測定し、得られた二次微分スペクトルの全測定波数における平均値および標準偏差(σ)を算出し、Kσ(K=3)を描いたが、3σを逸脱する正常品が幾らか見られたので、正常品が全く逸脱しなくなるまでKを変動させたところK=3.7以上で判別ラインを逸脱する正常品はなくなった(図3参照)。従って、シリコン被覆酸化亜鉛においては、3.7σを判別ライン(図3における太線)とした。
Example 3 Preparation of Discrimination Line for Silicon-Coated Zinc Oxide The near-infrared spectrum of 3 normal lots of silicon-coated zinc oxide was measured 12 times each, and the average value of the obtained second derivative spectra at all measured wave numbers and The standard deviation (σ) was calculated and Kσ (K = 3) was drawn. However, some normal products that deviated from 3σ were found. When K was varied until the normal products did not deviate at all, K = No more normal products that deviated from the discrimination line at 3.7 or higher (see Fig. 3). Accordingly, in the case of silicon-coated zinc oxide, 3.7σ is set as a discrimination line (thick line in FIG. 3).

実施例4 N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドについての異常品の判別
予め異常品とわかっているN-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドを1ロット用意し、その近赤外スペクトルを測定して、実施例1で作製した判別ラインと比較した。その結果を図4に示す。図4において、正常品の標準スペクトルに対する異常品の偏差の絶対値を実線、判別ラインを太線で示す。
Example 4 Discrimination of Abnormal Products for N- (3-Hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide N- (3-Hexadecyloxy- One lot of 2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide was prepared, its near infrared spectrum was measured, and compared with the discrimination line prepared in Example 1. The result is shown in FIG. In FIG. 4, the absolute value of the deviation of the abnormal product with respect to the standard spectrum of the normal product is indicated by a solid line, and the discrimination line is indicated by a bold line.

図4より、異常品は複数の波数において判別ラインを逸脱しており、異常品と正常品とを正しく判別できることがわかる。   From FIG. 4, it can be seen that the abnormal product deviates from the discrimination line at a plurality of wave numbers, and the abnormal product can be correctly discriminated from the normal product.

実施例5 ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液についての異常品の判別
実施例4と同様に異常品と判別ラインとを比較した。その結果を図5に示す。
Example 5 Discrimination of Abnormal Products for Poly (N-propanoylethyleneimine) Graft-Dimethylsiloxane / γ-Aminopropylmethylsiloxane Copolymer Monoethyl Sulfate Salts As in Example 4, the abnormal products were compared with the discrimination lines. The result is shown in FIG.

図5より、異常品は複数の波数において判別ラインを逸脱しており、異常品と正常品とを正しく判別できることがわかる。   As can be seen from FIG. 5, the abnormal product deviates from the discrimination line at a plurality of wave numbers, and the abnormal product and the normal product can be correctly discriminated.

実施例6 シリコン被覆酸化亜鉛についての異常品の判別
実施例4と同様に異常品と判別ラインとを比較した。その結果を図6に示す。
Example 6 Discrimination of Abnormal Products for Silicon-Coated Zinc Oxide Abnormal products and discrimination lines were compared in the same manner as in Example 4. The result is shown in FIG.

図6より、異常品は複数の波数において判別ラインを逸脱しており、異常品と正常品とを正しく判別できることがわかる。   From FIG. 6, it can be seen that the abnormal product deviates from the discrimination line at a plurality of wave numbers, and the abnormal product can be correctly discriminated from the normal product.

本発明は、製品製造工程の品質管理や原料受入れ試験の必要な産業に広く利用することができる。   The present invention can be widely used in industries that require quality control in product manufacturing processes and raw material acceptance tests.

N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドの正常品の近赤外スペクトルの偏差の絶対値および本発明により作製された判別ラインを示す図である。The figure which shows the absolute value of the deviation of the near-infrared spectrum of the normal product of N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide and the discrimination line prepared by the present invention It is. ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液の正常品の近赤外スペクトルの偏差の絶対値および本発明により作製された判別ラインを示す図である。The figure which shows the absolute value of the deviation of the near-infrared spectrum of a normal product of poly (N-propanoylethyleneimine) graft-dimethylsiloxane / γ-aminopropylmethylsiloxane copolymer monoethyl sulfate and the discrimination line prepared by the present invention It is. シリコン被覆酸化亜鉛の正常品の近赤外スペクトルの偏差の絶対値および本発明により作製された判別ラインを示す図である。It is a figure which shows the absolute value of the deviation of the near-infrared spectrum of the normal article of a silicon covering zinc oxide, and the discriminant line produced by the present invention. N-(3-ヘキサデシロキシ-2-ヒドロキシプロピル)-N-2-ヒドロキシエチルヘキサデカナミドの判別ラインおよび標準スペクトルと異常品の近赤外スペクトルとの偏差の絶対値を示す図である。It is a figure which shows the absolute value of the deviation of the discriminant line of N- (3-hexadecyloxy-2-hydroxypropyl) -N-2-hydroxyethylhexadecanamide and the standard spectrum and the near-infrared spectrum of an abnormal product. . ポリ(N-プロパノイルエチレンイミン)グラフト-ジメチルシロキサン/γ-アミノプロピルメチルシロキサンコポリマー硫酸モノエチル塩液の判別ラインおよび標準スペクトルと異常品の近赤外スペクトルとの偏差の絶対値を示す図である。FIG. 7 is a diagram showing a discrimination line of poly (N-propanoylethyleneimine) graft-dimethylsiloxane / γ-aminopropylmethylsiloxane copolymer monoethyl sulfate solution and an absolute value of deviation between a standard spectrum and a near-infrared spectrum of an abnormal product. . シリコン被覆酸化亜鉛の判別ラインおよび標準スペクトルと異常品の近赤外スペクトルとの偏差の絶対値を示す図である。It is a figure which shows the absolute value of the deviation of the discrimination | determination line and standard spectrum of a silicon coating zinc oxide, and the near-infrared spectrum of an abnormal article.

Claims (4)

a)正常品の近赤外スペクトルを測定する工程、
b)工程a)で得られた近赤外スペクトルをもとに標準スペクトルおよび標準偏差(σ)を算出する工程、ならびに
c)工程b)で得られた標準偏差(σ)をもとに判別ラインを作成する工程
を含む、異常品を判別するための判別ラインの作成方法。
a) a step of measuring a near-infrared spectrum of a normal product,
b) A step of calculating a standard spectrum and a standard deviation (σ) based on the near-infrared spectrum obtained in step a), and c) A discrimination based on the standard deviation (σ) obtained in step b). A method for creating a discrimination line for discriminating abnormal products, including a step of creating a line.
工程b)で使用される近赤外スペクトルが二次微分スペクトルである請求項1記載の判別ラインの作成方法。   The method for creating a discrimination line according to claim 1, wherein the near-infrared spectrum used in step b) is a second derivative spectrum. 工程c)で作成される判別ラインがKminσである請求項1または2記載の判別ラインの作成方法。 The method for creating a discrimination line according to claim 1 or 2, wherein the discrimination line created in step c) is K min σ. d)判別対象サンプルの近赤外スペクトルを測定する工程、および
e)請求項1の工程b)で得られた標準スペクトルに対する工程d)で得られた近赤外スペクトルとの偏差の絶対値と、請求項1〜3いずれか記載の方法により作成された判別ラインとを比較する工程
を含む、異常品の判別方法。
d) a step of measuring the near-infrared spectrum of the sample to be discriminated, and e) an absolute value of a deviation from the near-infrared spectrum obtained in step d) with respect to the standard spectrum obtained in step b) of claim 1. A method for discriminating abnormal products, comprising a step of comparing a discrimination line created by the method according to claim 1.
JP2004317434A 2004-11-01 2004-11-01 Discrimination method using near-infrared spectrum Pending JP2006126100A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2004317434A JP2006126100A (en) 2004-11-01 2004-11-01 Discrimination method using near-infrared spectrum

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2004317434A JP2006126100A (en) 2004-11-01 2004-11-01 Discrimination method using near-infrared spectrum

Publications (1)

Publication Number Publication Date
JP2006126100A true JP2006126100A (en) 2006-05-18

Family

ID=36720975

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2004317434A Pending JP2006126100A (en) 2004-11-01 2004-11-01 Discrimination method using near-infrared spectrum

Country Status (1)

Country Link
JP (1) JP2006126100A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011214940A (en) * 2010-03-31 2011-10-27 Caloria Japan Co Ltd Method and device for discriminating foreign matter contamination in object
KR101108671B1 (en) * 2009-12-11 2012-01-25 한국건설기술연구원 Identification Test Method for Fire Resistive Coatings in Near-Infrared Spectroscopy
CN106706558A (en) * 2017-01-10 2017-05-24 南京富岛信息工程有限公司 Method for eliminating abnormal sample in calibration set
CN114354537A (en) * 2022-01-14 2022-04-15 四川启睿克科技有限公司 Abnormal spectrum discrimination method based on American ginseng

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101108671B1 (en) * 2009-12-11 2012-01-25 한국건설기술연구원 Identification Test Method for Fire Resistive Coatings in Near-Infrared Spectroscopy
JP2011214940A (en) * 2010-03-31 2011-10-27 Caloria Japan Co Ltd Method and device for discriminating foreign matter contamination in object
CN106706558A (en) * 2017-01-10 2017-05-24 南京富岛信息工程有限公司 Method for eliminating abnormal sample in calibration set
CN106706558B (en) * 2017-01-10 2019-03-22 南京富岛信息工程有限公司 A method of rejecting calibration set exceptional sample
CN114354537A (en) * 2022-01-14 2022-04-15 四川启睿克科技有限公司 Abnormal spectrum discrimination method based on American ginseng

Similar Documents

Publication Publication Date Title
CN105466927B (en) Method for identifying, correcting and alarming abnormal reaction curve of turbidimetry
JP6912768B2 (en) Non-destructive inspection method and equipment for pear maturity by odor measurement
CN106932015B (en) Real-time characterization of coated drug tablets using raman spectroscopy
Schwanninger et al. Determination of lignin content in Norway spruce wood by Fourier transformed near infrared spectroscopy and partial least squares regression analysis. Part 2: Development and evaluation of the final model
EP3004839B1 (en) Method of and apparatus for correcting for intensity deviations in a spectrometer
Polak et al. Use of hyperspectral imaging for cake moisture and hardness prediction
CN110567937A (en) Competitive self-adaptive heavy-weighted key data extraction method for Raman spectrum analysis of insulating oil
US20130311136A1 (en) Rule-Based Sample Verification and Chemical Monitoring Methodology
JP2006126100A (en) Discrimination method using near-infrared spectrum
US20160209324A1 (en) Identifying presence of substances
CN104807398B (en) A kind of method and apparatus in OCD measurements for screening wave band
Likhachev Efficient thin-film stack characterization using parametric sensitivity analysis for spectroscopic ellipsometry in semiconductor device fabrication
EP4328685A1 (en) Process monitoring
Liu et al. Rapid detection of hydrolyzed leather protein adulteration in infant formula by near-infrared spectroscopy
JP6784756B2 (en) Arrangement to determine the achievable bond strength before connecting the materials continuously to the surface of the mating partner
JP2003232759A (en) Odor discrimination apparatus
JPH06300689A (en) Interior quality measuring method for unripe fruit by transmission method
US6671629B2 (en) Method and device for measuring characteristics of a sample
Westerhaus Eastern analytical symposium award for outstanding achievements in near infrared spectroscopy: my contributions to near infrared spectroscopy
JP6346991B2 (en) Method for extracting information contained in NMR measurement results as a signal
JP2008180618A (en) Surface defect detector
CN113029980B (en) Rapid nondestructive testing method for stability of sensory quality of tobacco sheet
RU2565356C1 (en) Method of separate measurement of mass fractions of oil and gas condensate in products of oil-gas condensate wells
SE516894C2 (en) Method for determining baking properties of flour
Nturambirwe et al. Performance of genetic algorithm in optimization of NIRS PLS models to predict apple fruit quality