JP2010210355A - Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy - Google Patents
Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy Download PDFInfo
- Publication number
- JP2010210355A JP2010210355A JP2009055779A JP2009055779A JP2010210355A JP 2010210355 A JP2010210355 A JP 2010210355A JP 2009055779 A JP2009055779 A JP 2009055779A JP 2009055779 A JP2009055779 A JP 2009055779A JP 2010210355 A JP2010210355 A JP 2010210355A
- Authority
- JP
- Japan
- Prior art keywords
- wavelength
- measurement
- light
- calibration curve
- absorbance
- 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
Links
- 238000005259 measurement Methods 0.000 title claims abstract description 93
- 238000000034 method Methods 0.000 title claims abstract description 61
- 235000013311 vegetables Nutrition 0.000 title claims abstract description 28
- 238000004497 NIR spectroscopy Methods 0.000 title abstract description 14
- 238000011088 calibration curve Methods 0.000 claims abstract description 79
- 238000002835 absorbance Methods 0.000 claims abstract description 32
- 238000000862 absorption spectrum Methods 0.000 claims abstract description 29
- 238000000691 measurement method Methods 0.000 claims abstract description 23
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 10
- 235000013305 food Nutrition 0.000 claims abstract description 10
- 235000013372 meat Nutrition 0.000 claims abstract description 7
- 230000001678 irradiating effect Effects 0.000 claims abstract description 4
- NHNBFGGVMKEFGY-UHFFFAOYSA-N Nitrate Chemical compound [O-][N+]([O-])=O NHNBFGGVMKEFGY-UHFFFAOYSA-N 0.000 claims description 28
- 238000011156 evaluation Methods 0.000 claims description 25
- 239000006185 dispersion Substances 0.000 claims description 19
- 235000009337 Spinacia oleracea Nutrition 0.000 claims description 17
- 235000003228 Lactuca sativa Nutrition 0.000 claims description 8
- 240000008415 Lactuca sativa Species 0.000 claims description 6
- 238000000611 regression analysis Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 230000001066 destructive effect Effects 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000000513 principal component analysis Methods 0.000 claims description 5
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 235000012045 salad Nutrition 0.000 claims description 4
- 235000007294 Brassica nipposinica Nutrition 0.000 claims description 2
- 240000007124 Brassica oleracea Species 0.000 claims description 2
- 235000003899 Brassica oleracea var acephala Nutrition 0.000 claims description 2
- 235000011301 Brassica oleracea var capitata Nutrition 0.000 claims description 2
- 235000001169 Brassica oleracea var oleracea Nutrition 0.000 claims description 2
- 235000010149 Brassica rapa subsp chinensis Nutrition 0.000 claims description 2
- 235000000536 Brassica rapa subsp pekinensis Nutrition 0.000 claims description 2
- 241000499436 Brassica rapa subsp. pekinensis Species 0.000 claims description 2
- 235000007516 Chrysanthemum Nutrition 0.000 claims description 2
- 244000189548 Chrysanthemum x morifolium Species 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims description 2
- 244000300264 Spinacia oleracea Species 0.000 claims 2
- 241000342995 Brassica rapa subsp. nipposinica Species 0.000 claims 1
- 229910002651 NO3 Inorganic materials 0.000 abstract description 6
- -1 nitrate ions Chemical class 0.000 abstract 1
- 241000219315 Spinacia Species 0.000 description 15
- 238000012628 principal component regression Methods 0.000 description 15
- 238000009826 distribution Methods 0.000 description 12
- 238000001228 spectrum Methods 0.000 description 9
- 238000010187 selection method Methods 0.000 description 7
- 241000208822 Lactuca Species 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000004611 spectroscopical analysis Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000003337 fertilizer Substances 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 150000002500 ions Chemical class 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000000491 multivariate analysis Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000007670 refining Methods 0.000 description 2
- 230000011514 reflex Effects 0.000 description 2
- 244000026811 Brassica nipposinica Species 0.000 description 1
- GRYLNZFGIOXLOG-UHFFFAOYSA-N Nitric acid Chemical compound O[N+]([O-])=O GRYLNZFGIOXLOG-UHFFFAOYSA-N 0.000 description 1
- 238000000540 analysis of variance Methods 0.000 description 1
- 238000004587 chromatography analysis Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000003673 groundwater Substances 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 238000004255 ion exchange chromatography Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910017604 nitric acid Inorganic materials 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 239000000618 nitrogen fertilizer Substances 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
- 150000003722 vitamin derivatives Chemical class 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
【課題】近赤外線分光法において、推定精度の高い検量線を得て、また、検量線の推定精度を低下させる不要な情報を削減して、目的成分濃度を高精度かつ迅速に非破壊計測する方法ならびに装置を提供する。
【解決手段】波長４００ｎｍ〜２５００ｎｍの範囲またはその一部範囲の波長光を測定対象の野菜、果物、肉類などの食物に照射し、その透過光及び／又は反射光を検出して吸光度スペクトルを取得し、測定全波長あるいは特定波長の吸光度から検量線を用いて測定対象の目的成分濃度を計測する非破壊計測法において、測定対象に対する波長光の照射範囲を所定領域に限定する。例えば、野菜内硝酸イオン濃度等の計測において、測定対象が株、葉、葉片と小さくなるにつれて計測精度が向上する。また、測定対象に照射すべき必要最小限の波長光を選択することができる。これにより、測定時間の短縮だけでなく、推定精度の高い検量線を得る。
【選択図】図１In near-infrared spectroscopy, a calibration curve with high estimation accuracy is obtained, and unnecessary information that reduces the estimation accuracy of the calibration curve is reduced, so that the target component concentration can be measured nondestructively with high accuracy. Methods and apparatus are provided.
An absorption spectrum is obtained by irradiating foods such as vegetables, fruits, and meats to be measured with light having a wavelength in a wavelength range of 400 nm to 2500 nm or a partial range thereof, and detecting transmitted light and / or reflected light. In the nondestructive measurement method in which the target component concentration of the measurement target is measured using the calibration curve from the absorbance of all the measured wavelengths or the specific wavelength, the irradiation range of the wavelength light on the measurement target is limited to a predetermined region. For example, in measuring the concentration of nitrate ions in vegetables, etc., the measurement accuracy improves as the measurement target becomes smaller, such as a stock, a leaf, and a leaf piece. Further, it is possible to select a minimum wavelength light to be irradiated to the measurement target. Thereby, not only the measurement time is shortened, but a calibration curve with high estimation accuracy is obtained.
[Selection] Figure 1
Description
本発明は、近赤外線分光法による野菜等の成分の非破壊計測法および非破壊計測装置に関するもので、特に、野菜中の硝酸イオン濃度の非破壊計測などに利用できる。 The present invention relates to a nondestructive measurement method and a nondestructive measurement apparatus for components such as vegetables by near infrared spectroscopy, and can be used particularly for nondestructive measurement of nitrate ion concentration in vegetables.
既に、ホウレンソウおよびレタスといった野菜について、これらの生葉を直接測定対象として、近赤外線分光法を使用することにより、野菜中の硝酸イオン濃度を非破壊で計測する技術が知られている（特許文献１）。かかる近赤外線分光法においては、測定条件を理想的な状態にして、波長分解能が高く波長範囲の広い高価な近赤外線分光計を使用することにより、推定精度の高い検量線を得ることができ、硝酸イオン濃度を高精度で非破壊計測することが可能である。
しかしながら、安価で波長分解能が低い分光計を使用し、かつ、現場レベルの外乱の多い条件で測定する場合は、推定精度の高い検量線を得ることが困難であり、硝酸イオン濃度を高精度で非破壊計測することができなかった。
For vegetables such as spinach and lettuce, a technique for measuring nitrate ion concentration in vegetables non-destructively by using near-infrared spectroscopy with these raw leaves as direct measurement targets is known (Patent Document 1). ). In such near-infrared spectroscopy, it is possible to obtain a calibration curve with high estimation accuracy by using an expensive near-infrared spectrometer with a high wavelength resolution and a wide wavelength range by making the measurement conditions ideal. It is possible to measure nitrate ion concentration with high accuracy and non-destructive measurement.
However, when using an inexpensive spectrometer with low wavelength resolution and measuring under conditions with a lot of disturbances at the field level, it is difficult to obtain a calibration curve with high estimation accuracy. Nondestructive measurement was not possible.
また、従来の近赤外線分光法においては、株や葉面内の部位別の硝酸濃度分布を考慮した測定は行われていなかった。近年、ハイパースペクトルカメラが開発され、２次元画像内の画素単位で可視・近赤外線スペクトルを計測することが可能になり、このカメラ計測により測定対象面内の濃度分布の測定が可能となった。
しかしながら、ハイパースペクトルカメラが取得する情報は非常に大きく処理時間がかかるため、推定精度を確保しつつ情報量の削減が要望されていた。
Moreover, in the conventional near-infrared spectroscopy, the measurement which considered the nitric acid concentration distribution according to the site | part in a strain | stump | stock or a leaf surface was not performed. In recent years, a hyperspectral camera has been developed, and it has become possible to measure a visible / near-infrared spectrum in units of pixels in a two-dimensional image, and this camera measurement has made it possible to measure the concentration distribution in the measurement target surface.
However, since the information acquired by the hyperspectral camera is very large and takes a long processing time, there has been a demand for reducing the amount of information while ensuring the estimation accuracy.
さらに、従来の近赤外線分光法においては、分光器を使用した後分光法を採用しており、分光器の必要性から装置のコスト低減を見込むことができない。より簡便で安価な計測装置を提供するためには前分光法が有効と言われている。前分光法の場合は特定の波長のみを使用することになる。現在半値幅の小さな単波長に近い光を照射可能なＬＥＤが安価に入手できるようになり、前分光法が現実的になってきた。
しかしながら、現状では、高い推定精度を保証する検量線に使用する特定の波長を抽出する手法が見当たらない。
Furthermore, the conventional near-infrared spectroscopy employs a post-spectrometry method using a spectroscope, and the cost of the apparatus cannot be expected due to the necessity of the spectroscope. In order to provide a simpler and cheaper measuring apparatus, it is said that pre-spectroscopy is effective. In the case of pre-spectroscopy, only a specific wavelength is used. LEDs that can irradiate light having a small half-value width and near a single wavelength can be obtained at low cost, and pre-spectroscopy has become practical.
However, at present, there is no method for extracting a specific wavelength used for a calibration curve that guarantees high estimation accuracy.
上記問題に鑑みて、本発明の第１の目的は、近赤外線分光法において、推定精度の高い検量線を得て、目的成分濃度を高精度で非破壊計測する方法ならびに装置を提供することである。また、本発明の第２の目的は、検量線の推定精度を低下させる不要な情報を削減して、目的成分濃度を高精度かつ迅速に非破壊計測する方法ならびに装置を提供することである。 In view of the above problems, a first object of the present invention is to provide a method and apparatus for obtaining a calibration curve with high estimation accuracy and nondestructive measurement of target component concentration with high accuracy in near infrared spectroscopy. is there. A second object of the present invention is to provide a method and apparatus for reducing the unnecessary information that lowers the estimation accuracy of the calibration curve and nondestructively measuring the target component concentration with high accuracy and speed.
本発明者らは、様々な検討を重ねた結果、本発明に係る非破壊計測法および非破壊計測装置を完成した。
すなわち、上記問題を解決すべく、本発明に係る第１の観点からは、
波長４００ｎｍ〜２５００ｎｍの範囲またはその一部範囲の波長光を測定対象の野菜、果物、肉類などの食物に照射し、その透過光及び／又は反射光を検出して吸光度スペクトルを取得し、測定全波長あるいは特定波長の吸光度から検量線を用いて測定対象の目的成分濃度を計測する非破壊計測法において、
測定対象に対する波長光の照射範囲を所定領域に限定することにより、線量線の推定精度を向上させることを特徴とする非破壊計測法が提供される。
As a result of various studies, the present inventors have completed the nondestructive measurement method and the nondestructive measurement apparatus according to the present invention.
That is, in order to solve the above problem, from the first viewpoint according to the present invention,
Irradiate foods such as vegetables, fruits, and meats to be measured with light in the wavelength range of 400 nm to 2500 nm, or a partial range thereof, detect the transmitted light and / or reflected light, obtain an absorbance spectrum, and measure all In the nondestructive measurement method that measures the concentration of the target component of the measurement object using the calibration curve from the absorbance of the wavelength or specific wavelength,
There is provided a nondestructive measurement method characterized by improving the estimation accuracy of the dose line by limiting the irradiation range of the wavelength light to the measurement object to a predetermined region.
測定対象部位の空間分解能を向上させ、測定範囲を可能な限り小さくすることにより、検量線の推定精度を向上できる。空間分解能とは測定対象の投影面積である。
近赤外線分光法を使用した野菜内硝酸イオン濃度等の計測において、測定対象が株、葉、葉片と小さくなるにつれて計測精度が向上することを、計測を繰り返す作業の中で、経験的に見出したものである。株、葉、葉片と小さくなるにつれて計測精度が向上する理由としては、株内または葉内の硝酸イオン濃度変動が大きいことが要因と推察している。
By improving the spatial resolution of the region to be measured and making the measurement range as small as possible, the estimation accuracy of the calibration curve can be improved. Spatial resolution is the projected area of the measurement object.
In the measurement of nitrate concentration in vegetables using near-infrared spectroscopy, we found empirically that the measurement accuracy improved as the measurement object became smaller, such as strains, leaves, leaf pieces, in the work of repeating measurement Is. The reason why the measurement accuracy is improved as the strain, leaf, and leaf piece become smaller is presumed to be due to a large fluctuation of nitrate ion concentration in the strain or leaf.
ここで、２次元画像内の画素単位で吸光度スペクトルを計測するカメラ計測法により、測定対象の投影面積である空間分解能を画素単位とすることが好ましい。
かかるカメラ計測法は、具体的には、ハイパースペクトルカメラ（ＨＳＣ）を用いて計測する。ハイパースペクトルカメラとは、ハイパースペクトル情報をもった画像を撮影するカメラである。ハイパースペクトル情報とは、光を波長としてとらえ、各波長における光の強度を測定したものである。
Here, it is preferable to set the spatial resolution, which is the projection area of the measurement target, in units of pixels by a camera measurement method that measures the absorbance spectrum in units of pixels in the two-dimensional image.
Specifically, this camera measurement method uses a hyperspectral camera (HSC) for measurement. A hyperspectral camera is a camera that captures an image having hyperspectral information. The hyperspectral information is obtained by measuring light as a wavelength and measuring the intensity of light at each wavelength.
また、上記の非破壊計測法における検量線は、下記の１）と２）の２つのステップによって得ることができる。
１）選択された所定の波長光による吸光度スペクトルを格納するデータ行列を特異値分解によりスコアとローディングとに分解し、目的成分の濃度の変動を要約する主要な成分を主成分分析によって抽出するステップ
２）説明変量をスコア、目的変量を目的成分の濃度とする重回帰分析を適用し、重回帰式を作成するステップ
A calibration curve in the above nondestructive measurement method can be obtained by the following two steps 1) and 2).
1) A step of decomposing a data matrix storing an absorbance spectrum of a selected light having a predetermined wavelength into a score and a loading by singular value decomposition and extracting principal components summarizing fluctuations in the concentration of a target component by principal component analysis 2) Step of creating a multiple regression equation by applying multiple regression analysis with the explanatory variable as the score and the target variable as the concentration of the target component
選択された所定の波長光による吸光度スペクトルに基づいて、検量線を作成することにより、検量線の推定精度を低下させる不要な情報を削減して、目的成分濃度を高精度かつ迅速に非破壊計測することが可能となる。 By creating a calibration curve based on the absorbance spectrum of the selected light of the specified wavelength, unnecessary information that reduces the estimation accuracy of the calibration curve can be reduced, and the target component concentration can be measured with high accuracy and speed. It becomes possible to do.
ここで、上記の所定の波長光の選択は、以下の（１）〜（６）の手順に従う方法で行う。
（１）測定全波長を測定波長とし、
（２）測定波長について波長別の吸光度分散を算出し、
（３）測定波長から最小分散を示す波長を削除し、
（４）残りの波長の吸光度を用いて検量線を作成し、
（５）評価データの相関係数を算出し、
（６）上記（２）〜（５）までを最小波長数（＝１）になるまで繰り返し、評価データの相関係数値が最も高いものを選択する。
Here, the selection of the predetermined wavelength light is performed by a method according to the following procedures (1) to (6).
(1) The measurement wavelength is the measurement wavelength,
(2) Calculate the absorbance dispersion by wavelength for the measurement wavelength,
(3) Delete the wavelength indicating the minimum dispersion from the measured wavelength,
(4) Create a calibration curve using the absorbance at the remaining wavelengths,
(5) Calculate the correlation coefficient of the evaluation data,
(6) The above (2) to (5) are repeated until the minimum number of wavelengths (= 1), and the one with the highest correlation coefficient value in the evaluation data is selected.
かかる方法によれば、非破壊計測法において、測定対象に照射すべき必要最小限の波長光を選択することができる。これにより、測定時間の短縮だけでなく、推定精度の高い検量線を得ることが可能となる。波長の中には推定精度の向上に障害になる波長が存在するため、高い推定精度を保証する検量線に使用する特定の波長を抽出するのである。 According to such a method, in the nondestructive measurement method, it is possible to select the minimum necessary wavelength light to be irradiated to the measurement target. This makes it possible to obtain a calibration curve with high estimation accuracy as well as shortening the measurement time. Among wavelengths, there are wavelengths that hinder the improvement of estimation accuracy, so a specific wavelength used for a calibration curve that guarantees high estimation accuracy is extracted.
上記の非破壊計測法において、特に好ましくは、波長６００ｎｍ〜１０００ｎｍの範囲またはその一部範囲の波長光を測定対象の野菜に照射する。
かかる波長光を測定対象の野菜に照射することにより、野菜中の硝酸イオン濃度の効率よく測定できる。この野菜は、具体的には、ホウレンソウ、サラダホウレンソウ、レタス、サニーレタス、サラダ菜、春菊、ターツァイ、チンゲンサイ、キャベツ、ハクサイ、コマツナ、及びミズナからなる群から選ばれる１種又は数種の野菜である。
In the above non-destructive measurement method, particularly preferably, the vegetable to be measured is irradiated with light having a wavelength in the range of 600 nm to 1000 nm or a partial range thereof.
By irradiating the vegetable to be measured with such wavelength light, the nitrate ion concentration in the vegetable can be measured efficiently. Specifically, this vegetable is one or several kinds of vegetables selected from the group consisting of spinach, salad spinach, lettuce, sunny lettuce, salad vegetables, spring chrysanthemum, tarzai, chingensai, cabbage, Chinese cabbage, komatsuna, and Mizuna. .
次に、本発明に係る第２の観点からは、
ａ）波長４００ｎｍ〜２５００ｎｍの範囲またはその一部範囲の波長光を測定対象の野菜、果物または肉類などの食物に照射する投光手段と、
ｂ）２次元画像内の画素単位で吸光度スペクトルを計測するカメラ手段と、
ｃ）測定対象に対する波長光の照射範囲を所定領域に限定して、前記カメラ手段を用いて測定対象の投影面積である空間分解能を画素単位とし、画素毎に得られた測定全波長あるいは特定波長の吸光度から検量線を作成する検量線作成手段と、
ｄ）空間分解能を画素単位とし、画素毎に得られた測定全波長あるいは特定波長の吸光度から、検量線を用いて測定対象の目的成分濃度を計測する成分解析手段と、
を少なくとも備えた非破壊計測装置が提供される。
Next, from the second viewpoint according to the present invention,
a) a light projecting means for irradiating food such as vegetables, fruits or meat to be measured with light having a wavelength in the wavelength range of 400 nm to 2500 nm or a partial range thereof;
b) camera means for measuring the absorbance spectrum in units of pixels in the two-dimensional image;
c) Limiting the irradiation range of the wavelength light to the measurement object to a predetermined area, using the camera means, the spatial resolution, which is the projection area of the measurement object, as a pixel unit, and measuring all wavelengths or specific wavelengths obtained for each pixel A calibration curve creating means for creating a calibration curve from the absorbance of
d) component analysis means for measuring the target component concentration of the measurement target using a calibration curve from the measured total wavelength or the absorbance at a specific wavelength obtained for each pixel with the spatial resolution as a pixel unit;
Is provided.
ここで、投光手段は、発光ダイオード（ＬＥＤ）で構成されていることが好ましい。現在半値幅の小さな単波長に近い光を照射可能なＬＥＤが安価に入手できることから、装置全体を安価に構成できるからである。また、カメラ手段は、ハイパースペクトルカメラ（ＨＳＣ）であることが好ましい。 Here, the light projecting means is preferably composed of a light emitting diode (LED). This is because an LED capable of emitting light close to a single wavelength with a small half-value width can be obtained at low cost, and the entire apparatus can be configured at low cost. The camera means is preferably a hyperspectral camera (HSC).
また、上記の非破壊計測装置における検量線作成手段は、プログラムであり、コンピュータに対して、
Ａ）所定の波長光を選択する手段と、
Ｂ）選択された所定の波長光による吸光度スペクトルを格納するデータ行列を特異値分解によりスコアとローディングとに分解し、目的成分の濃度の変動を要約する主要な成分を主成分分析によって抽出する抽出手段と、
Ｃ）説明変量をスコア、目的変量を目的成分の濃度とする重回帰分析を適用し、重回帰式を作成する作成手段として、
機能させるためのプログラムである。
Further, the calibration curve creating means in the nondestructive measuring apparatus is a program,
A) means for selecting light of a predetermined wavelength;
B) Extraction that extracts the main component that summarizes the variation of the concentration of the target component by principal component analysis by decomposing the data matrix storing the absorbance spectrum of the selected light of the predetermined wavelength into the score and loading by singular value decomposition Means,
C) As a creation means for creating a multiple regression equation by applying multiple regression analysis with the explanatory variable as a score and the target variable as the concentration of the target component,
It is a program to make it function.
ここで、上記の所定の波長光を選択する手段は、プログラムであり、コンピュータに対して、
（１）測定全波長を測定波長とする手順、
（２）測定波長について波長別の吸光度分散を算出する手順、
（３）測定波長から最小分散を示す波長を削除する手順、
（４）残りの波長の吸光度を用いて検量線を作成する手順、
（５）評価データの相関係数を算出する手順、
（６）上記（２）〜（５）までを最小波長数（＝１）になるまで繰り返し、評価データの相関係数値が最も高いものを選択する手順、
を実行させるためのプログラムである。
Here, the means for selecting the predetermined wavelength light is a program, and for the computer,
(1) Procedure for setting all measurement wavelengths as measurement wavelengths,
(2) Procedure for calculating absorbance dispersion for each wavelength for the measurement wavelength;
(3) Procedure for deleting the wavelength indicating the minimum dispersion from the measured wavelength;
(4) Procedure for creating a calibration curve using the absorbance of the remaining wavelengths,
(5) a procedure for calculating a correlation coefficient of evaluation data;
(6) A procedure for selecting the one having the highest correlation coefficient value in the evaluation data by repeating the steps (2) to (5) until the minimum number of wavelengths (= 1) is reached.
Is a program for executing
また、本発明に係る第３の観点からは、
波長４００ｎｍ〜２５００ｎｍの範囲またはその一部範囲の波長光を測定対象の野菜、果物、肉類などの食物に照射し、その透過光及び／又は反射光を検出して吸光度スペクトルを取得し、測定全波長あるいは特定波長の吸光度から検量線を用いて測定対象の目的成分濃度を計測する非破壊計測法における検量線を作成するためのプログラムであって、
コンピュータに
（１）測定全波長を測定波長とする手順、
（２）測定波長について波長別の吸光度分散を算出する手順、
（３）測定波長から最小分散を示す波長を削除する手順、
（４）残りの波長の吸光度を用いて検量線を作成する手順、
（５）評価データの相関係数を算出する手順、
（６）上記（２）〜（５）までを最小波長数（＝１）になるまで繰り返し、評価データの相関係数値が最も高いものを選択する手順、
を実行させるためのプログラムを記録したコンピュータ読み取り可能な記録媒体が提供される。
From the third viewpoint according to the present invention,
Irradiate foods such as vegetables, fruits, and meats to be measured with light in the wavelength range of 400 nm to 2500 nm, or a partial range thereof, detect the transmitted light and / or reflected light, obtain an absorbance spectrum, and measure all A program for creating a calibration curve in a non-destructive measurement method that measures the concentration of a target component of a measurement object using a calibration curve from absorbance at a wavelength or a specific wavelength,
(1) Procedure for setting all the wavelengths to be measured to the computer,
(2) Procedure for calculating absorbance dispersion for each wavelength for the measurement wavelength;
(3) Procedure for deleting the wavelength indicating the minimum dispersion from the measured wavelength;
(4) Procedure for creating a calibration curve using the absorbance of the remaining wavelengths,
(5) a procedure for calculating a correlation coefficient of evaluation data;
(6) A procedure for selecting the one having the highest correlation coefficient value in the evaluation data by repeating the steps (2) to (5) until the minimum number of wavelengths (= 1) is reached.
A computer-readable recording medium on which a program for executing is recorded is provided.
上述したように、本発明によれば、近赤外線分光法において、推定精度の高い検量線を得て、目的成分濃度を高精度で非破壊計測できるといった効果を有する。
また、検量線の推定精度を低下させる不要な情報を削減して、目的成分濃度を高精度かつ迅速に非破壊計測できるといった効果を有する。
As described above, according to the present invention, in near-infrared spectroscopy, a calibration curve with high estimation accuracy can be obtained, and the target component concentration can be measured nondestructively with high accuracy.
Moreover, unnecessary information that reduces the estimation accuracy of the calibration curve can be reduced, and the target component concentration can be measured with high accuracy and speed in a nondestructive manner.
以下、本発明の実施の形態について、野菜中の硝酸イオン濃度を非破壊で計測する装置を例に挙げ、図面を参照しながら詳細に説明していく。なお、本発明の範囲は、以下の実施例や図示例に限定されるものではなく、幾多の変更及び変形が可能である。 Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings, taking as an example an apparatus for measuring the nitrate ion concentration in vegetables nondestructively. The scope of the present invention is not limited to the following examples and illustrated examples, and various changes and modifications can be made.
実施例１では、ハイパースペクトルカメラを使用して硝酸イオン濃度分布計測を行うための検量線を作成した場合に、高い推定精度が得られる点について詳細に説明する。
実施例１の非破壊計測装置は、図１に示すように、波長６００ｎｍ〜１０００ｎｍの範囲の波長光を測定対象に照射するレフランプ１と、２次元画像内の画素単位で吸光度スペクトルを計測するハイパースペクトルカメラ２と、コンピュータ３から構成される。
コンピュータ３には、検量線作成プログラムと成分解析プログラムが搭載されている。検量線作成プログラムは、測定対象に対する波長光の照射範囲を所定領域に限定して、ハイパースペクトルカメラ２を用いて測定対象の投影面積である空間分解能を画素単位とし、画素毎に得られた特定波長の吸光度から検量線を作成する機能を有する。また、成分解析プログラムは、画素毎に得られた特定波長の吸光度から検量線を用いて測定対象の目的成分濃度を計測する機能を有する。
図１は、レフランプ１から近赤外線波長の波長光を供試植物の葉４（ホウレンソウ、コマツナ）に照射し、その照射光６に伴う反射光７をハイパースペクトルカメラ２で捕らえて波長毎の画像を撮影する。ハイパースペクトルカメラ２はＬＡＮ等の通信ケーブルでコンピュータ３と接続されている。
なお、ハイパースペクトルカメラ２とコンピュータ３の間のデータは、有線または無線の通信ネットワーク、あるいは、メモリ媒体などでデータを授受してもよい。またコンピュータ３は、携帯電話やＰＤＡ（Personal Digital Assistant）などの携帯情報端末でもよい。
The first embodiment will explain in detail that a high estimation accuracy can be obtained when a calibration curve for performing nitrate ion concentration distribution measurement using a hyperspectral camera is created.
As shown in FIG. 1, the nondestructive measuring apparatus according to the first embodiment includes a reflex lamp 1 that irradiates a measurement target with light having a wavelength ranging from 600 nm to 1000 nm, and a hyper that measures an absorbance spectrum in units of pixels in a two-dimensional image. It consists of a spectrum camera 2 and a computer 3.
The computer 3 is loaded with a calibration curve creation program and a component analysis program. The calibration curve creation program limits the irradiation range of the wavelength light to the measurement target to a predetermined area, and uses the hyperspectral camera 2 to set the spatial resolution, which is the projection area of the measurement target, as a pixel unit, and to obtain a specific obtained for each pixel It has a function of creating a calibration curve from the absorbance of the wavelength. In addition, the component analysis program has a function of measuring the target component concentration of the measurement target using the calibration curve from the absorbance of the specific wavelength obtained for each pixel.
FIG. 1 shows a wavelength of near-infrared wavelength emitted from a reflex lamp 1 to a leaf 4 (spinach, komatsuna) of a test plant, and the reflected light 7 associated with the irradiated light 6 is captured by the hyperspectral camera 2 for each wavelength. Shoot. The hyperspectral camera 2 is connected to the computer 3 via a communication cable such as a LAN.
The data between the hyperspectral camera 2 and the computer 3 may be exchanged via a wired or wireless communication network or a memory medium. The computer 3 may be a mobile information terminal such as a mobile phone or a PDA (Personal Digital Assistant).
この検量線作成プログラムが検量線を作成する手順について、硝酸イオン濃度を算出する場合を例に挙げて説明する。
（１）吸光スペクトルデータの入力
ハイパースペクトルカメラによる吸光スペクトルデータを入力する。
A procedure for creating a calibration curve by the calibration curve creation program will be described by taking as an example the case of calculating a nitrate ion concentration.
(1) Input of absorption spectrum data Input absorption spectrum data from a hyperspectral camera.
（２）吸光スペクトルデータの前処理
吸光スペクトルデータから多変量データを算出する。６０７〜９６７ｎｍまでの波長範囲を使用する。分解能が９ｎｍであるからｍ個（ｍ ＝４１）の波長における吸光度が１サンプル分格納された１行×ｍ列のベクトルデータxが得られることになる。この多変量データに前処理を施す。例えば各波長別に（各列毎に）標準化変換したり、ベースライン（ゼロ点）移動の影響を回避するため中心化処理をしたり、あるいは１次微分や２次微分等の処理をする。これらの前処理は測定対象の性質や測定の目的に応じて適宜選択される。本実施例では、サンプル平均をサンプル値から差し引いた値を標準偏差で除す標準化変換を行う。
(2) Preprocessing of absorption spectrum data Multivariate data is calculated from the absorption spectrum data. A wavelength range from 607 to 967 nm is used. Since the resolution is 9 nm, 1 row × m column vector data x in which absorbances at m wavelengths (m = 41) are stored for one sample is obtained. Preprocessing is performed on the multivariate data. For example, standardization conversion is performed for each wavelength (for each column), centralization processing is performed to avoid the influence of baseline (zero point) movement, or processing such as primary differentiation and secondary differentiation is performed. These pretreatments are appropriately selected according to the properties of the measurement object and the purpose of the measurement. In this embodiment, standardization conversion is performed by dividing the value obtained by subtracting the sample average from the sample value by the standard deviation.
（３）主成分回帰分析（ＰＣＲ）処理
次に、多変量解析を行う。多変量解析では一般に主成分分析と重回帰分析を共用した主成分回帰分析（ＰＣＲ）法やＰＬＳ（Ｐａｒｔｉａｌ
Ｌｅａｓｔ Ｓｑｕａｒｅｓ）法を使用する。本実施例ではＰＣＲ法を用いる。なお、ＰＣＲ法については、特許文献１の段落００６９に詳細に説明されているので、ここでは説明を省略する。
(3) Principal component regression analysis (PCR) processing Next, multivariate analysis is performed. In multivariate analysis, Principal component regression analysis (PCR) and PLS (Partial) generally share principal component analysis and multiple regression analysis.
The Last Squares method is used. In this example, the PCR method is used. Since the PCR method is described in detail in paragraph 0069 of Patent Document 1, description thereof is omitted here.
（４）主成分数の決定処理
スペクトルデータの主要な変動を捕らえているのは第何主成分までかという問題は検量線の精度には重要である。主成分の数を過剰に多く取ると推定誤差が大きくなるからである。本実施例では主成分の数を最大で３０に限定する。なお、主成分数については、特許文献１の段落００７０に詳細に説明されているので、ここでは説明を省略する。
(4) Determination of the number of principal components The problem of how many principal components are capturing the main fluctuations in the spectral data is important for the accuracy of the calibration curve. This is because an excessively large number of principal components increases the estimation error. In this embodiment, the number of main components is limited to 30 at the maximum. Note that the number of principal components is described in detail in paragraph 0070 of Patent Document 1, and thus the description thereof is omitted here.
（５）重回帰式の作成処理
主成分数を決定したらスコア行列で該当する列のみ切りだして重回帰分析を行う。最終的に偏回帰係数、標準偏回帰係数、回帰式の分散分析、寄与率（決定係数）、回帰係数の検定結果並びに回帰推定値と実測値のデータが出力される。
(5) Multiple regression equation creation processing Once the number of principal components is determined, only the corresponding column is extracted from the score matrix and multiple regression analysis is performed. Finally, partial regression coefficient, standard partial regression coefficient, analysis of variance of regression equation, contribution rate (determination coefficient), test result of regression coefficient, and data of regression estimation value and actual measurement value are output.
（６）回帰ベクトルの作成処理
最終的に、各波長の吸光度ベクトルとの内積により濃度推定値を与える回帰ベクトルを計算する。硝酸イオン濃度（ｙ）は、y ＝ ｘＢの式から求められる。このＢ（ｍ行×１列）の要素は各波長に対応する吸光度にかける偏回帰係数であり、回帰ベクトルから決定されるものである。
(6) Regression Vector Creation Process Finally, a regression vector that gives an estimated concentration value is calculated by the inner product with the absorbance vector of each wavelength. The nitrate ion concentration (y) is obtained from the equation y = xB. The element of B (m rows × 1 column) is a partial regression coefficient to be applied to the absorbance corresponding to each wavelength, and is determined from the regression vector.
かかる検量線作成プログラムは、計測時に実行することができる。計測時に実行する場合は、吸光スペクトルを濃度分布が等しくなるように校正用データと評価用データとに２分割し、吸光スペクトルデータとして校正用データを用いる。なお、予め計測前に検量線作成プログラムを実行して、検量線を作成してもかまわない。 Such a calibration curve creation program can be executed during measurement. When executed at the time of measurement, the absorption spectrum is divided into two parts for calibration data and evaluation data so that the concentration distribution is equal, and the calibration data is used as the absorption spectrum data. Note that a calibration curve may be created in advance by executing a calibration curve creation program before measurement.
本実施例１の装置を用いて、検量線を作成したものを例示する。図２はホウレンソウの検量線の作成例であり、図３はコマツナの検量線の作成例である。
また、本実施例１の装置を用いて作成した硝酸イオン濃度葉面分布例を示す。図４はホウレンソウの一葉の面内の硝酸イオン濃度葉面分布例であり、図５はコマツナの一葉の面内の硝酸イオン濃度葉面分布例である。
図４から、ホウレンソウの場合は、硝酸イオン濃度が葉の葉脈部分と葉柄の部分で高くなっていることが確認できる。また図５から、コマツナの場合は、硝酸イオン濃度が葉全体に高くなっていることが確認できる。
図４や図５に示されるように、狭い葉面内でも予想以上に大きな濃度変動を持つことが確認できる。
なお、実際に、イオンクロマト法による破壊計測を使用して、葉を小片に分割して部位別に濃度計測をすると、同様に大きな濃度変動が確認できている。
The thing which produced the calibration curve using the apparatus of the present Example 1 is illustrated. FIG. 2 is an example of creating a calibration curve for spinach, and FIG. 3 is an example of creating a calibration curve for Komatsuna.
Moreover, the nitrate ion density | concentration leaf surface distribution example produced using the apparatus of the present Example 1 is shown. FIG. 4 is an example of leaf surface distribution of nitrate ion concentration in one leaf of spinach, and FIG. 5 is an example of leaf surface distribution of nitrate ion concentration in one leaf of Komatsuna.
From FIG. 4, in the case of spinach, it can be confirmed that the nitrate ion concentration is high in the leaf vein portion and the petiole portion. Moreover, from FIG. 5, in the case of Komatsuna, it can be confirmed that the nitrate ion concentration is high throughout the leaves.
As shown in FIG. 4 and FIG. 5, it can be confirmed that the density fluctuation is larger than expected even in a narrow leaf surface.
In fact, when the disruption measurement by the ion chromatography method is used and the leaf is divided into small pieces and the concentration is measured for each part, a large concentration fluctuation can be confirmed in the same manner.
従来の分光計は受光部が1つしかなく、株や葉の広い範囲に光を照射して反射光や透過光を受光しているため、一株や一葉で一つのスペクトルしか得られない。大きなサンプルになると部位により光の当たり方に差異が生じ、サンプルの一部のみの情報を含む光を受光することになる。このためスペクトルは大きな濃度変動を持つ株や葉の一部分の平均的な性質を表すことになる。
しかしながら、従来の破壊法によって得られる濃度実測値は、株や葉全体を抽出して得られるため、検量線作成に必要なスペクトルと実測値の両者間で面内の測定部位が異なることなる。このことが原因で検量線の精度が向上しなかったのである。
A conventional spectrometer has only one light receiving unit, and irradiates light over a wide area of a stock or leaf to receive reflected light or transmitted light. Therefore, only one spectrum can be obtained per stock or leaf. When a large sample is used, a difference occurs in how the light strikes depending on the part, and light including information on only a part of the sample is received. For this reason, the spectrum represents the average properties of a part of a strain or leaf having large concentration fluctuations.
However, since the measured concentration value obtained by the conventional disruption method is obtained by extracting the entire strain and leaves, the in-plane measurement site differs between the spectrum and the measured value necessary for preparing the calibration curve. For this reason, the accuracy of the calibration curve was not improved.
一方、本実施例のように、ハイパースペクトルカメラを使用した場合、スペクトル測定部位は画像中の画素単位で確定可能でありスペクトルと実測値の両者は面内の測定部位に差が生じにくくなる。このように測定部位を詳細に把握する、即ち空間分解能を詳細化してスペクトルと実測値を準備することにより、検量線の推定精度を向上させるのである。 On the other hand, when a hyperspectral camera is used as in the present embodiment, the spectrum measurement site can be determined in units of pixels in the image, and the difference between the spectrum and the actual measurement value is less likely to occur in the in-plane measurement site. Thus, the accuracy of the calibration curve estimation is improved by grasping the measurement site in detail, that is, by preparing the spectrum and the actual measurement value by refining the spatial resolution.
すなわち、波長分解能よりも、空間分解能を精細化することが検量線精度の向上により有効なのである。測定部位の空間分解能は画素単位レベルまで詳細化することが好ましく、スペクトルと実測値を準備する。画素単位レベルまで詳細化することで、検量線の推定精度が更に向上する。
以上、説明したように、近赤外線分光法による野菜内硝酸イオン濃度等の非破壊計測法において、測定部位の空間分解能を向上し、測定範囲を可能な限り小さくすることで、検量線の推定精度が向上できることが理解できよう。
That is, it is more effective to improve the accuracy of the calibration curve to make the spatial resolution finer than the wavelength resolution. The spatial resolution of the measurement site is preferably detailed to the pixel unit level, and a spectrum and an actual measurement value are prepared. Refinement to the pixel unit level further improves the estimation accuracy of the calibration curve.
As explained above, in nondestructive measurement methods such as nitrate concentration in vegetables by near infrared spectroscopy, the accuracy of calibration curve estimation is improved by improving the spatial resolution of the measurement site and making the measurement range as small as possible. It can be understood that can be improved.
以下に、本実施例の方法を用いて、ホウレンソウとコマツナの硝酸イオン濃度を測定した結果について説明する。
実際にホウレンソウ１６８サンプル、コマツナ１６０サンプルを使用して検量線を作成したところ、推定値と実測値の相関係数はホウレンソウで0.933476、コマツナで0.883287となった。
Below, the result of having measured the nitrate ion concentration of spinach and Komatsuna using the method of the present embodiment will be described.
When a calibration curve was actually created using 168 samples of spinach and 160 samples of Komatsuna, the correlation coefficient between the estimated value and the actually measured value was 0.933476 for spinach and 0.883287 for Komatsuna.
推定精度の比較のために、ハイパースペクトルカメラを使用しないで、一般に使用されている近赤外線分光計（ＦＡＮＴＥＣ社製ＦＲＵＩＴ ＱＵＡＬＩＴＹ
ＡＮＡＬＹＺＥＲ ６００〜１１００ｎｍ、分解能２ｎｍ）を使用して、レタスの株全体およびコマツナ葉中の硝酸イオン濃度を測定する検量線を作成したものを準備した。
一般に使用されている近赤外線分光計を使用したものでは、測定対象が株全体の場合では、推定値と実測値との相関係数は0.775549、一葉の場合では0.737068に留まり、実用化としては困難な推定精度であった。
For comparison of estimation accuracy, the near-infrared spectrometer (FRUIT QUALITY manufactured by FANTEC) is generally used without using a hyperspectral camera.
(Analyzer 600-1100 nm, resolution 2 nm) was used to prepare a calibration curve for measuring the nitrate ion concentration in the whole lettuce strain and Komatsuna leaves.
When using a commonly used near-infrared spectrometer, the correlation coefficient between the estimated value and the actual measurement value is 0.775549 when the measurement target is the entire stock, and 0.737068 in the case of one leaf, which is difficult to put into practical use. The estimation accuracy was high.
これに対して、ハイパースペクトルカメラを使用した本実施例の方法においては、図２および図３に示すように、測定対象が一葉の場合で推定値と実測値との相関係数（ＰＬＳ法）は、ホウレンソウで0.933476、コマツナで0.883287となった。特に、ホウレンソウの場合は、相関係数が０．９を超えており、実用化できる精度であることが理解できる。
また、この値から、本実施例の方法により得られた推定精度は実用可能な精度であることが確認できる。
On the other hand, in the method of the present embodiment using a hyperspectral camera, as shown in FIGS. 2 and 3, the correlation coefficient (PLS method) between the estimated value and the actually measured value when the measurement object is a single leaf. Was 0.933476 for spinach and 0.883287 for Komatsuna. In particular, in the case of spinach, the correlation coefficient exceeds 0.9, and it can be understood that the accuracy is practical.
Further, from this value, it can be confirmed that the estimation accuracy obtained by the method of the present embodiment is a practical accuracy.
さらに、注目すべきは、ハイパースペクトルカメラの波長分解能は９nmに対して、これまで使用してきた分光計では２ｎｍであり、ハイパースペクトルカメラの方が、波長分解能が粗いにもかかわらず推定精度が向上していることである。
このことからも波長分解能よりは空間分解能を精細化することが検量線精度の向上に有効であることがわかる。なお、従来の分光計を使用しても照射範囲を小さく限定することにより同様な推定精度の向上が見込まれる。
以上説明した如く、近赤外線分光法を使用した非破壊計測装置には、その計測部に検量線が必要であるものの、これまで検量線の推定精度が余り高くなかった。本実施例の方法を使用して検量線を作成すれば、精度の高い検量線が得られることになる。
Furthermore, it should be noted that the wavelength resolution of the hyperspectral camera is 9 nm, whereas the spectrometer used so far is 2 nm. The estimation accuracy of the hyperspectral camera is improved despite the coarser wavelength resolution. Is.
From this, it can be seen that refining the spatial resolution rather than the wavelength resolution is effective in improving the accuracy of the calibration curve. Even if a conventional spectrometer is used, the same estimation accuracy can be improved by limiting the irradiation range to be small.
As described above, a nondestructive measuring apparatus using near infrared spectroscopy requires a calibration curve in its measuring unit, but the accuracy of estimation of the calibration curve has not been so high so far. If a calibration curve is created using the method of this embodiment, a calibration curve with high accuracy can be obtained.
実施例２では、検量線の推定精度を低下させる不要な波長光を削減して、目的成分濃度を高精度かつ迅速に非破壊計測するための、波長光の選択方法について詳細に説明する。
波長光の選択方法は、図６に示すフローに従って行う。
先ず、始めに、吸光スペクトルデータを入力する。そして、設定された波長分解能を入力し、測定波長範囲と波長分解能から波長数を算出する。ハイパースペクトルカメラの分解能１ｎｍの場合は、可視光〜近赤外光（６００〜１０００ｎｍ）の合計４０１個の波長を用いてハイパースペクトルカメラにより画像撮影する。測定波長範囲の全波長の吸光度を用いて検量線を作成し、評価データの相関係数を算出する。ここまでは、従来の検量線の作成と同じである。
（ステップＳ１）測定全波長を測定波長とする。
ここでは、初期値として測定波長範囲の測定可能な波長を用いる。ハイパースペクトルカメラの分解能１ｎｍの場合は、測定波長は、可視光〜近赤外光（６００〜１０００ｎｍ）の合計４０１個の波長となる。
（ステップＳ２）測定波長について波長別の吸光度分散を算出する。
図７の吸光スペクトルと分散グラフに示すように、測定対象サンプルが異なれば、吸光スペクトルも異なる。この吸光スペクトルの分散度合いを表す波長別の吸光度分散を算出する。
（ステップＳ３）測定波長から最小分散を示す波長を削除する。
波長数から１減算される。
（ステップＳ４）残りの波長の吸光度を用いて検量線を作成する。
（ステップＳ５）評価データの相関係数を算出する。
（ステップＳ６）上記のステップＳ２〜Ｓ５までを最小波長数（＝１）になるまで繰り返し、評価データの相関係数値が最も高いものを選択する。
In the second embodiment, a wavelength light selection method for reducing the unnecessary wavelength light that reduces the estimation accuracy of the calibration curve and performing nondestructive measurement of the target component concentration with high accuracy and speed will be described in detail.
The wavelength light selection method is performed according to the flow shown in FIG.
First, absorption spectrum data is input. Then, the set wavelength resolution is input, and the number of wavelengths is calculated from the measurement wavelength range and the wavelength resolution. When the resolution of the hyperspectral camera is 1 nm, images are captured by the hyperspectral camera using a total of 401 wavelengths from visible light to near infrared light (600 to 1000 nm). A calibration curve is created using the absorbance at all wavelengths in the measurement wavelength range, and the correlation coefficient of the evaluation data is calculated. Up to this point, it is the same as the creation of a conventional calibration curve.
(Step S <b> 1) All measurement wavelengths are set as measurement wavelengths.
Here, a measurable wavelength in the measurement wavelength range is used as an initial value. When the resolution of the hyperspectral camera is 1 nm, the measurement wavelength is a total of 401 wavelengths from visible light to near infrared light (600 to 1000 nm).
(Step S2) Absorbance dispersion for each wavelength is calculated for the measurement wavelength.
As shown in the absorption spectrum and the dispersion graph of FIG. 7, the absorption spectrum is different when the sample to be measured is different. The absorbance dispersion for each wavelength representing the degree of dispersion of the absorbance spectrum is calculated.
(Step S3) The wavelength indicating the minimum dispersion is deleted from the measured wavelength.
One is subtracted from the number of wavelengths.
(Step S4) A calibration curve is created using the absorbance at the remaining wavelengths.
(Step S5) The correlation coefficient of the evaluation data is calculated.
(Step S6) The above steps S2 to S5 are repeated until the minimum number of wavelengths (= 1), and the one with the highest correlation coefficient value in the evaluation data is selected.
従来の近赤外線分光法を使用した非破壊計測では、可視光〜近赤外光（６００〜１０００ｎｍ）の非常に多くの波長を使用する。ハイパースペクトルカメラの分解能１ｎｍの場合は、合計４０１個の波長を用いて画像撮影することとなる。
本実施例２では、上記実施例１の非破壊計測装置を使用し、あらかじめ小数の波長光を限定してハイパースペクトルカメラで画像撮影し、吸光スペクトルを得ることで、測定時間の短縮だけでなく、推定精度の高い検量線を得ることができる点について以下に説明する。
In nondestructive measurement using conventional near infrared spectroscopy, a very large number of wavelengths from visible light to near infrared light (600 to 1000 nm) are used. When the resolution of the hyperspectral camera is 1 nm, images are captured using a total of 401 wavelengths.
In the second embodiment, not only the measurement time is shortened by using the non-destructive measuring apparatus of the first embodiment, capturing images with a hyperspectral camera by limiting a small number of wavelengths in advance, and obtaining an absorption spectrum. The point that a calibration curve with high estimation accuracy can be obtained will be described below.
先ず、あらかじめ小数の波長光を限定すべく、波長を選択する方法について説明する。
近赤外線分光法における波長選択の方法は、具体的には、始めに校正用データについて等間隔に波長を間引く（削除する）。そして、残った波長から波長別吸光度分散が小さい波長を一つずつ間引いて（削除して）、その都度、残りの波長を使用して検量線を作成する。これを、残りの波長が２波長になるまでこの操作を繰り返す。また、この等間隔間引きが本来の分解能の２０分の１になるまで繰り返す。
ここで、波長別吸光度分散が小さい波長を一つずつ削除する処理は、具体的には、目的成分の濃度変動に敏感な順に波長をソートしていき、鈍感な波長を１波長ずつ削除する。
そして、得られた検量線の精度は、評価用データを代入して得られる実測値の相関係数により評価する。一番高い評価が得られた検量線で使用した波長が目的成分の濃度推定に必要最小限の波長となる。
First, a method for selecting a wavelength in order to limit a small number of wavelength lights in advance will be described.
Specifically, in the wavelength selection method in the near-infrared spectroscopy, first, the wavelength is thinned out (deleted) at equal intervals for the calibration data. Then, one wavelength is thinned out (deleted) one by one from the remaining wavelengths, and a calibration curve is created using the remaining wavelengths each time. This operation is repeated until the remaining wavelengths become two wavelengths. Further, this equal interval thinning is repeated until it becomes 1/20 of the original resolution.
Here, the process of deleting each wavelength having a small absorbance dispersion for each wavelength is specifically performed by sorting the wavelengths in order of sensitivity to the concentration fluctuation of the target component, and deleting the insensitive wavelengths one by one.
Then, the accuracy of the obtained calibration curve is evaluated by the correlation coefficient of the actually measured values obtained by substituting the evaluation data. The wavelength used in the calibration curve with the highest evaluation is the minimum wavelength necessary for estimating the concentration of the target component.
本実施例では、検量線の作成に、主成分回帰分析（ＰＣＲ）法およびＰＬＳ（Ｐａｒｔｉａｌ Ｌｅａｓｔ Ｓｑｕａｒｅｓ）法の２通り行った。以下データを示しながら、本実施例２の方法により、ＰＣＲおよびＰＬＳ共に計測精度が向上することを、図を示しながら説明する。
測定対象は、神戸大学農学部内にある圃場と温室で栽培したレタスである。サンプルは株を1単位とし、全１６１サンプルを使用した。ＦＡＮＴＥＣ社製のＦＲＵＩＴ ＱＵＡＬＩＴＹ ＡＮＡＬＹＺＥＲを使用して、６００〜１１００ｎｍまでの波領域で株全体の近赤外線吸光スペクトルを測定し、吸光度を算出した。精算時間は１０、２０、３０、４０、５０ｍｓである。
In this example, the calibration curve was created in two ways: a principal component regression analysis (PCR) method and a PLS (Partial Last Squares) method. Hereinafter, it will be described with reference to the drawings that the measurement accuracy of both PCR and PLS is improved by the method of the second embodiment while showing data.
The object of measurement is lettuce cultivated in a field and greenhouse in the Faculty of Agriculture, Kobe University. The sample was a strain, and a total of 161 samples were used. Using FRUIT QUALITY ANALYZER manufactured by FANTEC, the near-infrared absorption spectrum of the entire strain was measured in the wave region from 600 to 1100 nm, and the absorbance was calculated. The settlement time is 10, 20, 30, 40, 50 ms.
従来法による硝酸イオン濃度測定にはイオンクロマト法を用いた。従来法の計測には、東亜ＤＫＫ（株）製イオン分析計ＩＡ−３００を使用した。これら吸光度スペクトルと硝酸イオン濃度実測値のデータセットは、検正用データ１０８、評価用データ５３に分割した。 The ion chromatographic method was used for measuring the nitrate ion concentration by the conventional method. For the measurement of the conventional method, an ion analyzer IA-300 manufactured by Toa DKK Co., Ltd. was used. The data set of the absorbance spectrum and the measured nitrate ion concentration was divided into calibration data 108 and evaluation data 53.
図８にＰＣＲ法を使用して作成した検量線の推定精度の結果を、図９にＰＬＳ法を使用して作成した検量線の推定精度の結果を示す。
また、図１０にＰＣＲ法における推定精度の変化を、図１１にＰＬＳ法における推定精度の変化を示す。図１０のＰＣＲ法の場合は、測定全波長（４０１波長）を使用するものと、本実施例の波長選択方法を用いて選択された２８１波長を使用するものを比較している。また、図１１のＰＬＳ法の場合は、測定全波長（４０１波長）を使用するものと、本実施例の波長選択方法を用いて選択された１１１波長を使用するものを比較している。
図８では、分解能１ｎｍのときに、最も推定精度が高くなったのは波長数が２８１のときで、評価データ相関係数は0.788907であったことが示されている。また、分解能６ｎｍのときに、最も推定精度が高くなったのは波長数が４９のときで、評価データ相関係数は0.787021であったことが示されている。全波長を使用した場合の評価データ相関係数は0.778945であり、ＰＣＲ法において、小数の波長のみを使用した方が、推定精度が高くなった。
一方、図９では、分解能１ｎｍのときに、最も推定精度が高くなったのは波長数が１１１のときで、評価データ相関係数は0.729719であったことが示されている。また、分解能３ｎｍのときに、最も推定精度が高くなったのは波長数が３７のときで、評価データ相関係数は0.724824であったことが示されている。全波長を使用した場合の評価データ相関係数は0.604671であり、ＰＬＳ法において、小数の波長のみを使用した方が、推定精度が高くなった。
以上のことから、波長の中には推定精度の向上に障害になる波長が存在することが明らかになった。全波長を使用しなくても高い推定精度が得られることが理解できよう。
FIG. 8 shows the results of calibration curve estimation accuracy created using the PCR method, and FIG. 9 shows the results of calibration curve estimation accuracy created using the PLS method.
FIG. 10 shows a change in estimation accuracy in the PCR method, and FIG. 11 shows a change in estimation accuracy in the PLS method. In the case of the PCR method of FIG. 10, a comparison is made between those using all the measured wavelengths (401 wavelengths) and those using the 281 wavelengths selected using the wavelength selection method of this embodiment. Further, in the case of the PLS method of FIG. 11, a comparison is made between those using all measured wavelengths (401 wavelengths) and those using 111 wavelengths selected using the wavelength selection method of this embodiment.
FIG. 8 shows that when the resolution is 1 nm, the estimation accuracy is highest when the number of wavelengths is 281 and the evaluation data correlation coefficient is 0.788907. Further, it is shown that when the resolution is 6 nm, the estimation accuracy is highest when the number of wavelengths is 49, and the evaluation data correlation coefficient is 0.787021. The evaluation data correlation coefficient when all wavelengths are used is 0.778945, and in the PCR method, the estimation accuracy is higher when only a small number of wavelengths are used.
On the other hand, FIG. 9 shows that when the resolution is 1 nm, the estimation accuracy is highest when the number of wavelengths is 111 and the evaluation data correlation coefficient is 0.729719. Further, it is shown that when the resolution is 3 nm, the estimation accuracy is highest when the number of wavelengths is 37, and the evaluation data correlation coefficient is 0.724824. When all wavelengths are used, the evaluation data correlation coefficient is 0.604671. In the PLS method, the estimation accuracy is higher when only a few wavelengths are used.
From the above, it has become clear that there are wavelengths that hinder the improvement of estimation accuracy among wavelengths. It will be understood that high estimation accuracy can be obtained without using all wavelengths.
以上説明した実施形態の装置や方法は、硝酸イオン濃度のみならず果物の糖度計測やビタミン量の計測器にも応用できる。携帯式の装置であれば、野菜の生産、流通、販売の現場で計測使用できる。
また、食品の硝酸イオン濃度が高いと人体に有害である。農産物の生産現場では窒素肥料の施肥管理が精密に行うことができ、低硝酸野菜が可能になり商品価値を上げることができると同時に余剰肥料の低減により、生産コストの低減と余剰窒素の地下水への流出による水質汚濁防止に役立つ。本発明に係る方法や装置を用いることで、流通、販売の現場において、専門家でなくても簡便に硝酸イオン濃度を測定できるようになり、低硝酸野菜を選別が可能になり商品価値の差別化に寄与できる。
The apparatus and method of the embodiment described above can be applied not only to nitrate ion concentration but also to measuring sugar content of fruits and measuring instrument of vitamin amount. If it is a portable device, it can be used in the production, distribution and sale of vegetables.
Moreover, if the nitrate ion concentration of food is high, it is harmful to the human body. At the production site of agricultural products, the fertilizer management of nitrogen fertilizer can be performed precisely, enabling low nitrate vegetables to increase product value, and at the same time, reducing surplus fertilizer to reduce production costs and surplus nitrogen groundwater Helps prevent water pollution caused by spills. By using the method and apparatus according to the present invention, it becomes possible to easily measure the nitrate ion concentration even in the field of distribution and sales, even if it is not an expert, making it possible to select low nitrate vegetables and distinguishing commercial value Can contribute to
本発明は、ホウレンソウおよびレタスといった野菜等の食品の生産、流通、販売、消費の各過程における食品品質管理方法や装置として有用である。 INDUSTRIAL APPLICABILITY The present invention is useful as a food quality management method and apparatus in each process of production, distribution, sale, and consumption of foods such as spinach and lettuce.
１ レフランプ
２ ハイパースペクトルカメラ
３ コンピュータ
４ 供試植物の葉
５ 通信ケーブル
６ 照射光
７ 反射光
DESCRIPTION OF SYMBOLS 1 Ref lamp 2 Hyper spectrum camera 3 Computer 4 Leaf of test plant 5 Communication cable 6 Irradiation light 7 Reflected light
Claims (13)
測定対象に対する波長光の照射範囲を所定領域に限定することにより、線量線の推定精度を向上させることを特徴とする非破壊計測法。 Irradiate foods such as vegetables, fruits, and meats to be measured with light in the wavelength range of 400 nm to 2500 nm, or a partial range thereof, detect the transmitted light and / or reflected light, obtain an absorbance spectrum, and measure all In the nondestructive measurement method that measures the concentration of the target component of the measurement object using the calibration curve from the absorbance of the wavelength or specific wavelength,
A nondestructive measurement method characterized by improving the estimation accuracy of a dose line by limiting the irradiation range of wavelength light to a measurement object to a predetermined region.
（１）選択された所定の波長光による吸光度スペクトルを格納するデータ行列を特異値分解によりスコアとローディングとに分解し、目的成分の濃度の変動を要約する主要な成分を主成分分析によって抽出するステップと、
（２）説明変量をスコア、目的変量を目的成分の濃度とする重回帰分析を適用し、重回帰式を作成するステップと、
によって得られることを特徴とする請求項１に記載の非破壊計測法。 The calibration curve is
(1) A data matrix storing an absorbance spectrum of a selected light having a predetermined wavelength is decomposed into a score and a loading by singular value decomposition, and main components summarizing fluctuations in the concentration of the target component are extracted by principal component analysis. Steps,
(2) applying multiple regression analysis with the explanatory variable as a score and the target variable as the concentration of the target component, and creating a multiple regression equation;
The nondestructive measurement method according to claim 1, wherein
（１）測定全波長を測定波長とし、
（２）測定波長について波長別の吸光度分散を算出し、
（３）測定波長から最小分散を示す波長を削除し、
（４）残りの波長の吸光度を用いて検量線を作成し、
（５）評価データの相関係数を算出し、
（６）上記（２）〜（５）までを最小波長数になるまで繰り返し、評価データの相関係数値が最も高いものを選択することを特徴とする請求項４に記載の非破壊計測法。 The selection of the predetermined wavelength light is as follows:
(1) The measurement wavelength is the measurement wavelength,
(2) Calculate the absorbance dispersion by wavelength for the measurement wavelength,
(3) Delete the wavelength indicating the minimum dispersion from the measured wavelength,
(4) Create a calibration curve using the absorbance at the remaining wavelengths,
(5) Calculate the correlation coefficient of the evaluation data,
(6) The nondestructive measurement method according to claim 4, wherein the above (2) to (5) are repeated until the minimum number of wavelengths is reached, and the one having the highest correlation coefficient value in the evaluation data is selected.
２次元画像内の画素単位で吸光度スペクトルを計測するカメラ手段と、
測定対象に対する波長光の照射範囲を所定領域に限定して、前記カメラ手段を用いて測定対象の投影面積である空間分解能を画素単位とし、画素毎に得られた測定全波長あるいは特定波長の吸光度から検量線を作成する検量線作成手段と、
空間分解能を画素単位とし、画素毎に得られた測定全波長あるいは特定波長の吸光度から、検量線を用いて測定対象の目的成分濃度を計測する成分解析手段と、
を少なくとも備えたことを特徴とする非破壊計測装置。 A light projecting means for irradiating food such as vegetables, fruits or meat to be measured with light having a wavelength in the range of 400 nm to 2500 nm or a partial range thereof;
Camera means for measuring an absorbance spectrum in units of pixels in a two-dimensional image;
Limiting the irradiation range of the wavelength light to the measurement target to a predetermined area, using the camera means, the spatial resolution, which is the projected area of the measurement target, in units of pixels, and the measured total wavelength or specific wavelength absorbance obtained for each pixel A calibration curve creation means for creating a calibration curve from
Component analysis means for measuring the target component concentration of the measurement target using a calibration curve from the absorbance of the measured total wavelength or specific wavelength obtained for each pixel, with a spatial resolution as a pixel unit,
A nondestructive measuring device characterized by comprising at least.
コンピュータに
所定の波長光を選択する手段と、
選択された所定の波長光による吸光度スペクトルを格納するデータ行列を特異値分解によりスコアとローディングとに分解し、目的成分の濃度の変動を要約する主要な成分を主成分分析によって抽出する抽出手段と、
説明変量をスコア、目的変量を目的成分の濃度とする重回帰分析を適用し、重回帰式を作成する作成手段として、
機能させるためのプログラムである、
ことを特徴とする請求項８に記載の非破壊計測装置。 The calibration curve creating means is a program,
Means for selecting light of a predetermined wavelength for the computer;
An extraction means for decomposing a data matrix storing an absorbance spectrum of a selected light having a predetermined wavelength into a score and a loading by singular value decomposition and extracting principal components by summarizing fluctuations in the concentration of the target component by principal component analysis; ,
As a creation means to create a multiple regression equation by applying multiple regression analysis with the explanatory variable as the score and the target variable as the concentration of the target component,
It is a program to make it function,
The nondestructive measuring apparatus according to claim 8.
コンピュータに
（１）測定全波長を測定波長とする手順、
（２）測定波長について波長別の吸光度分散を算出する手順、
（３）測定波長から最小分散を示す波長を削除する手順、
（４）残りの波長の吸光度を用いて検量線を作成する手順、
（５）評価データの相関係数を算出する手順、
（６）上記（２）〜（５）までを最小波長数（＝１）になるまで繰り返し、評価データの相関係数値が最も高いものを選択する手順、
を実行させるためのプログラムである、
ことを特徴とする請求項１１に記載の非破壊計測装置。 The means for selecting the predetermined wavelength light is a program,
(1) Procedure for setting all the wavelengths to be measured to the computer,
(2) Procedure for calculating absorbance dispersion for each wavelength for the measurement wavelength;
(3) Procedure for deleting the wavelength indicating the minimum dispersion from the measured wavelength;
(4) Procedure for creating a calibration curve using the absorbance of the remaining wavelengths,
(5) a procedure for calculating a correlation coefficient of evaluation data;
(6) A procedure for selecting the one having the highest correlation coefficient value in the evaluation data by repeating the steps (2) to (5) until the minimum number of wavelengths (= 1) is reached.
Is a program for executing
The nondestructive measuring apparatus according to claim 11.
前記検量線を作成するためのプログラムであって、
コンピュータに
（１）測定全波長を測定波長とする手順、
（２）測定波長について波長別の吸光度分散を算出する手順、
（３）測定波長から最小分散を示す波長を削除する手順、
（４）残りの波長の吸光度を用いて検量線を作成する手順、
（５）評価データの相関係数を算出する手順、
（６）上記（２）〜（５）までを最小波長数になるまで繰り返し、評価データの相関係数値が最も高いものを選択する手順、
を実行させるためのプログラムを記録したコンピュータ読み取り可能な記録媒体。
Irradiate foods such as vegetables, fruits, and meats to be measured with light in the wavelength range of 400 nm to 2500 nm, or a partial range thereof, detect the transmitted light and / or reflected light, obtain an absorbance spectrum, and measure all In the nondestructive measurement method that measures the concentration of the target component of the measurement object using the calibration curve from the absorbance of the wavelength or specific wavelength,
A program for creating the calibration curve,
(1) Procedure for setting all the wavelengths to be measured to the computer,
(2) Procedure for calculating absorbance dispersion for each wavelength for the measurement wavelength;
(3) Procedure for deleting the wavelength indicating the minimum dispersion from the measured wavelength;
(4) Procedure for creating a calibration curve using the absorbance of the remaining wavelengths,
(5) a procedure for calculating a correlation coefficient of evaluation data;
(6) The above steps (2) to (5) are repeated until the minimum number of wavelengths is reached, and a procedure for selecting the one having the highest correlation coefficient value in the evaluation data,
The computer-readable recording medium which recorded the program for performing this.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2009055779A JP2010210355A (en) | 2009-03-09 | 2009-03-09 | Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2009055779A JP2010210355A (en) | 2009-03-09 | 2009-03-09 | Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JP2010210355A true JP2010210355A (en) | 2010-09-24 |
Family
ID=42970690
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2009055779A Pending JP2010210355A (en) | 2009-03-09 | 2009-03-09 | Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP2010210355A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012014327A1 (en) * | 2010-07-30 | 2012-02-02 | トヨタ自動車 株式会社 | Spectrum measuring device for mobile object and spectrum measuring method for mobile object |
| KR101158474B1 (en) | 2010-10-19 | 2012-06-20 | 한국식품연구원 | Methods for Analyzing Insoluble Ingredients in Salt Using Near Infrared |
| CN102564964A (en) * | 2011-12-29 | 2012-07-11 | 南京林业大学 | Spectral image-based meat quality visual non-contact detection method |
| CN102628800A (en) * | 2012-03-26 | 2012-08-08 | 北京农业智能装备技术研究中心 | Method and system for measuring plant leaf chlorophyll content |
| WO2012109516A1 (en) * | 2011-02-10 | 2012-08-16 | University Of Delaware | Methods and systems for identifying a point of interest on the periphery of an object |
| JP2015007621A (en) * | 2013-05-31 | 2015-01-15 | キヤノン株式会社 | Data processing method and device |
| CN105717051A (en) * | 2016-04-22 | 2016-06-29 | 合肥美菱股份有限公司 | System capable of rapidly detecting fruit and vegetable freshness and refrigerator |
| JP2018136142A (en) * | 2017-02-20 | 2018-08-30 | 株式会社トプコン | Measurement device, measurement method, and control program for measurement device |
| JP2019101040A (en) * | 2017-12-07 | 2019-06-24 | 国立研究開発法人農業・食品産業技術総合研究機構 | Method, device, and program for nondestructive measurement of nitrate ion concentration |
| CN113125459A (en) * | 2021-04-16 | 2021-07-16 | 合肥泰禾智能科技集团股份有限公司 | Improve reflector structure of fruit surface detection camera uniformity of picking up picture |
| JP2022008658A (en) * | 2017-02-20 | 2022-01-13 | 株式会社トプコン | Measuring device, measuring method and control program of measuring device |
| JP2022190227A (en) * | 2021-06-14 | 2022-12-26 | ハカルプラス株式会社 | Method and device for measuring nitrate nitrogen concentration |
| US11724286B2 (en) | 2013-11-01 | 2023-08-15 | Tomra Sorting Nv | Method and apparatus for detecting matter |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08320287A (en) * | 1995-05-24 | 1996-12-03 | Satake Eng Co Ltd | Spectroscopic measurement device |
| JP2001318053A (en) * | 2000-05-10 | 2001-11-16 | Shizuoka Prefecture | Plant quality evaluation device, breeding selection method and quality evaluation method |
| JP2003121351A (en) * | 2001-10-09 | 2003-04-23 | National Agricultural Research Organization | Meat freshness determination method |
| JP2003527594A (en) * | 2000-03-13 | 2003-09-16 | オートライン インコーポレイテッド | Apparatus and method for measuring and correlating fruit properties with visible / near infrared spectra |
| WO2005111583A1 (en) * | 2004-05-17 | 2005-11-24 | The New Industry Research Organization | Method for nondestructively examining component of vegetable or the like by near-infrared spectroscopy and its device |
| JP2006250827A (en) * | 2005-03-11 | 2006-09-21 | Pasuko:Kk | Analytical method for growth condition of crop |
-
2009
- 2009-03-09 JP JP2009055779A patent/JP2010210355A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08320287A (en) * | 1995-05-24 | 1996-12-03 | Satake Eng Co Ltd | Spectroscopic measurement device |
| JP2003527594A (en) * | 2000-03-13 | 2003-09-16 | オートライン インコーポレイテッド | Apparatus and method for measuring and correlating fruit properties with visible / near infrared spectra |
| JP2001318053A (en) * | 2000-05-10 | 2001-11-16 | Shizuoka Prefecture | Plant quality evaluation device, breeding selection method and quality evaluation method |
| JP2003121351A (en) * | 2001-10-09 | 2003-04-23 | National Agricultural Research Organization | Meat freshness determination method |
| WO2005111583A1 (en) * | 2004-05-17 | 2005-11-24 | The New Industry Research Organization | Method for nondestructively examining component of vegetable or the like by near-infrared spectroscopy and its device |
| JP2006250827A (en) * | 2005-03-11 | 2006-09-21 | Pasuko:Kk | Analytical method for growth condition of crop |
Non-Patent Citations (1)
| Title |
|---|
| 蔦 瑞樹 他: ""ハイパースペクトルシステムによる近赤外分光イメージング手法〜メロン糖分布の可視化事例〜"", 映像情報メディア学会誌, vol. Volume 56, Number 12, JPN6013020691, 1 December 2002 (2002-12-01), JP, pages 2037 - 2040, ISSN: 0002521718 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5321735B2 (en) * | 2010-07-30 | 2013-10-23 | トヨタ自動車株式会社 | Spectrum measuring apparatus for moving body and spectrum measuring method for moving body |
| US9121818B2 (en) | 2010-07-30 | 2015-09-01 | Toyota Jidosha Kabushiki Kaisha | Movable body spectrum measuring apparatus and movable body spectrum measuring method |
| WO2012014327A1 (en) * | 2010-07-30 | 2012-02-02 | トヨタ自動車 株式会社 | Spectrum measuring device for mobile object and spectrum measuring method for mobile object |
| KR101158474B1 (en) | 2010-10-19 | 2012-06-20 | 한국식품연구원 | Methods for Analyzing Insoluble Ingredients in Salt Using Near Infrared |
| WO2012109516A1 (en) * | 2011-02-10 | 2012-08-16 | University Of Delaware | Methods and systems for identifying a point of interest on the periphery of an object |
| CN102564964A (en) * | 2011-12-29 | 2012-07-11 | 南京林业大学 | Spectral image-based meat quality visual non-contact detection method |
| CN102628800A (en) * | 2012-03-26 | 2012-08-08 | 北京农业智能装备技术研究中心 | Method and system for measuring plant leaf chlorophyll content |
| JP2015007621A (en) * | 2013-05-31 | 2015-01-15 | キヤノン株式会社 | Data processing method and device |
| US11724286B2 (en) | 2013-11-01 | 2023-08-15 | Tomra Sorting Nv | Method and apparatus for detecting matter |
| CN105717051A (en) * | 2016-04-22 | 2016-06-29 | 合肥美菱股份有限公司 | System capable of rapidly detecting fruit and vegetable freshness and refrigerator |
| JP2022008658A (en) * | 2017-02-20 | 2022-01-13 | 株式会社トプコン | Measuring device, measuring method and control program of measuring device |
| JP2018136142A (en) * | 2017-02-20 | 2018-08-30 | 株式会社トプコン | Measurement device, measurement method, and control program for measurement device |
| JP7331058B2 (en) | 2017-02-20 | 2023-08-22 | 株式会社トプコン | Measuring device, measuring method and measuring device control program |
| JP2019101040A (en) * | 2017-12-07 | 2019-06-24 | 国立研究開発法人農業・食品産業技術総合研究機構 | Method, device, and program for nondestructive measurement of nitrate ion concentration |
| JP7169643B2 (en) | 2017-12-07 | 2022-11-11 | 国立研究開発法人農業・食品産業技術総合研究機構 | Nitrate ion concentration nondestructive measuring method, nitrate ion concentration nondestructive measuring device, and nitrate ion concentration nondestructive measuring program |
| CN113125459A (en) * | 2021-04-16 | 2021-07-16 | 合肥泰禾智能科技集团股份有限公司 | Improve reflector structure of fruit surface detection camera uniformity of picking up picture |
| JP2022190227A (en) * | 2021-06-14 | 2022-12-26 | ハカルプラス株式会社 | Method and device for measuring nitrate nitrogen concentration |
| JP7679937B2 (en) | 2021-06-14 | 2025-05-20 | ハカルプラス株式会社 | Method and device for measuring nitrate nitrogen concentration |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2010210355A (en) | Method and apparatus for nondestructive measurement of component of vegetable etc. using near-infrared spectroscopy | |
| Zhang et al. | Non-destructive prediction of soluble solids and dry matter contents in eight apple cultivars using near-infrared spectroscopy | |
| Wu et al. | Prediction of beef quality attributes using VIS/NIR hyperspectral scattering imaging technique | |
| Li et al. | Evaluating the performance of a consumer scale SCiO™ molecular sensor to predict quality of horticultural products | |
| Xu et al. | Rapid prediction and visualization of moisture content in single cucumber (Cucumis sativus L.) seed using hyperspectral imaging technology | |
| Sun et al. | How to predict the sugariness and hardness of melons: A near-infrared hyperspectral imaging method | |
| Wang et al. | Technology using near infrared spectroscopic and multivariate analysis to determine the soluble solids content of citrus fruit | |
| Peirs et al. | Effect of natural variability among apples on the accuracy of VIS-NIR calibration models for optimal harvest date predictions | |
| Peirs et al. | Effect of biological variability on the robustness of NIR models for soluble solids content of apples | |
| Cao et al. | Soluble solids content and pH prediction and varieties discrimination of grapes based on visible–near infrared spectroscopy | |
| Urraca et al. | Estimation of total soluble solids in grape berries using a hand‐held NIR spectrometer under field conditions | |
| JP5441703B2 (en) | Visible / Near-Infrared Spectroscopy and Grape Brewing Method | |
| Lin et al. | Determination of grain protein content by near-infrared spectrometry and multivariate calibration in barley | |
| Andersen et al. | The potential of laser-induced breakdown spectroscopy for industrial at-line monitoring of calcium content in comminuted poultry meat | |
| Tang et al. | Multispectral imaging for predicting sugar content of ‘Fuji’apples | |
| Li et al. | Pears characteristics (soluble solids content and firmness prediction, varieties) testing methods based on visible-near infrared hyperspectral imaging | |
| de Carvalho et al. | Rapid detection of whey in milk powder samples by spectrophotometric and multivariate calibration | |
| Huang et al. | Evaluating quality of tomato during storage using fusion information of computer vision and electronic nose | |
| JP5604763B2 (en) | Transmission near-infrared spectroscopy measurement apparatus and transmission near-infrared spectroscopy measurement method | |
| Liu et al. | Near-infrared prediction of edible oil frying times based on Bayesian Ridge Regression | |
| Bai et al. | Laser light backscattering image to predict moisture content of mango slices with different ripeness during drying process | |
| Miao et al. | Non-destructive determination of ginsenosides in ginseng by combined hyperspectral and X-ray techniques based on ensemble learning | |
| Bai et al. | Prediction of moisture content of Agaricus bisporus slices as affected by vacuum freeze drying using hyperspectral imaging | |
| Wang et al. | Measurement of SSC in processing tomatoes (Lycopersicon esculentum Mill.) by applying Vis‐NIR hyperspectral transmittance imaging and multi‐parameter compensation models | |
| Tirado-Kulieva et al. | Predicting the evolution of pH and total soluble solids during coffee fermentation using near-infrared spectroscopy coupled with chemometrics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20111201 |
|
| A977 | Report on retrieval |
Free format text: JAPANESE INTERMEDIATE CODE: A971007 Effective date: 20130419 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20130501 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20130628 |
|
| A02 | Decision of refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A02 Effective date: 20131120 |