JP4697764B2

JP4697764B2 - Method for judging the quality of gel-forming foods

Info

Publication number: JP4697764B2
Application number: JP2001301653A
Authority: JP
Inventors: 東一郎高井; 英明河原; 一男細谷; 郁夫富樫; 正人西; 原成天野; 隆弘松浦
Original assignee: Takai Tofu and Soymilk Equipment Co
Current assignee: Takai Tofu and Soymilk Equipment Co
Priority date: 2001-09-28
Filing date: 2001-09-28
Publication date: 2011-06-08
Anticipated expiration: 2021-09-28
Also published as: JP2003106995A

Description

【０００１】
【発明の属する技術分野】
本発明は例えば、豆腐をはじめ、卵豆腐、蒲鉾、チーズ等のゲル状食品や、加工や保存中に凝固又はゲル化する性質を有する豆乳、生卵、すり身、牛乳等の液体又はゾル状の食品ゲル形成性食品について、分光分析的な計測によって製品内部の品質を非破壊的又は被接触的に判定する方法および二次元又は三次元位置における品質判定を行う装置に関するものである。
【０００２】
【従来の技術】
食品製造工場の品質管理では、一般に、中間製品又は最終製品の品質について品質管理担当者が生産ロット毎や定期的に行う抜き取り検査では試験装置や化学分析法を用いて客観的な判断を下せるが、検査結果が得られるまで時間がかかる問題や検査漏れの問題がある。また、生産中は作業者が全品について注意を払い、目視や手触りによって良否を判断しているが、不注意や判断ミスは払拭できない。
【０００３】
ゲル形成性食品は、例えば豆乳に凝固剤を加え凝固させて豆腐にするように、加熱・冷却・凝固剤添加・発酵・酵素反応等の処理によって液状又はゾル状から固体状又はゲル状に変化させる食品である。一般食品と同様に、製品の抜き取り検査によって、一角の製品を開封してレオメーター等の物性試験機を用いてゲル強度などを評価している。破壊試験であるためその分がロスとなる。さらに万一不良品が検出されれば、たとえ不良品が一部分であっても、該当ロット全てを廃棄しなければならない。ゲル化のように、物性の変化を伴う加工作業や品質判断は作業者の主観、経験と勘に大きく依存している。ゲル状になった中間状態や中間製品は包装前に不良品を見分けることは容易であるが、自動化されたラインでは何らかの原因でムラや欠損等のある不良製品が出荷されてしまう場合がある。プリンやヨーグルト等の市販製品では、カップ上部に空間を設けてゲル化した状態を目視判断しやすくするようにしているが、空気による酸化や微生物の発生や型崩れ等の問題もある。また充填豆腐や卵豆腐のように、包装後にゲル化する場合、一見、そのゲル化の良否を判別しにくく、手触り等の感覚しか頼りにならない場合がある。外観上の異常はないが、内部に異常がある場合、不良品を見落としてしまうこともある。希に凝固不十分の製品が出荷され、販売先や消費者からの苦情処理や商品回収、さらには損害賠償責任問題まで発生し、その食品製造者にとって多大な負担を強いられることになる。
【０００４】
ゲル形成性食品の品質は、一般に形状（大きさ、欠損、重量等）、外観（色彩、重量、光沢、きめ等）、食感（硬さ、弾力、保水性）、味などの品質項目が挙げられる。一部は測定機器を用いて客観的に測定されるが、味や食感などは人が味わって判断しているのが現状である。
【０００５】
また、豆乳や牛乳等の液状又はゾル状食品（飲料）の場合、製品は液体又はゾル状態が正常であるが、何らかの原因で、凝集物や沈殿物・浮遊物、粘性増加やゲル化など、不良品となる場合がある。
【０００６】
製品の表面形状検査の目的では、例えば、包装された豆腐の欠損等について、蛍光灯を用いたスリット光の照射とＣＣＤカメラで画像処理によって判定する方法（特開平０８―２１７０２９）や、スポット投光器および受光センサーを多数配置することによって判定する方法（特開平１０−６２１３２）が開示されている。これらの方法では、単に外形の形状のみを計測する方法であるが、容器の色やフィルム上の印字の影響を受けるため、十分ではない。
【０００７】
包装前の豆腐等に関しては、一般的な光切断法を用いて、欠損を判定する方法（特開２００１−１３３２３３）が開示されているが、包装された製品では適用できない。
【０００８】
液状薬品又は食品の凝固判定装置としてＸ線を用いる画像解析装置の開示（登録第２９３１８３８号や特開２００１−２４２１００）がある。これらの装置は被対象物に気泡を含んでいる必要があり、前者では被対象物を振動させて、分散した気泡の状態を測定する方法である。いずれも液状の食品には適用できるが、ゲル状食品には適用できない。また放射線を扱うため、安全管理が必要になる。
【０００９】
可視光の分光分析法を用いて被対象物の成分を評価する方法として、例えば、果実等の被対象物の透過又は反射光を分割し、複数の特定波長フィルターを通過させ、それら吸光度の差から内部品質、例えば熟度を評価する方法（特開昭５２−８３２７４）が開示されている。
【００１０】
近赤外線分光分析法による有機物の品質評価方法（近赤外分析法）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、食品業界ではアミロース等の含有量を測定する米食味計（特公平１−４９８９０）や糖質を測定する果物糖度計（特許２５１７８５８、特開平１−３０１１４７）等が実用化されている。また、ココア含有食品のｐＨ（水素イオンの含量）を評価する方法（特開２００１−１７０８４）、食品の加工前後における成分の違いを測定し、加工状態を評価する方法（特開平７−１２７２２、特開平７−１０７９２５）が開示されている。ただし、豆腐類をはじめ、ゲル形成性食品に、可視光や近赤外線・赤外線による分光分析法を適用した例はない。
【００１１】
また、青果物では近赤外線領域の波長の吸収を利用して、硬さ（熟度）を予測するシステムも考案されている（特開平１−３０１１４７）が、青果物の熟成は一般に多くの化学成分の変化を伴っている。炊飯処理後の米飯を直接測定し、外観、食味の他、硬さ、粘りという物性値を評価する方法（特開平７−２７０３１２）が開示されているが、炊飯条件や米の性質によって吸収水分が異なり、一部の成分変化を伴っている。
【００１２】
いずれの方法も、アミロースや水素イオン濃度等の成分含量の変化に基づいた分析手法である。硬さや粘りという物性値は成分値との間接的相関に基づいて評価されている。そのため、成分が同一かつ変化がなく、物性が異なる場合は品質の判別が難しい面があった。これまでは、近赤外分光分析法は被測定物の化学成分量の違いを非破壊計測する目的に研究・実用化されてきた経緯がある。しかし、ゲル形成性食品について可視光や近赤外線等を使った分光分析法を適用し、その内部品質を判定した事例は未だ開示されていない。また、物体の三次元測定方法として、光切断法を用いる方法や特定波長域の光を用いる光切断法は物体の外形を画像で捉える場合に使われる公知の技術で、物体の表面的な欠陥、欠損、歪みの検出や外形測定に専ら用いられていた。被対象物内部の二次元又は三次元部位における性状判定に光切断法を応用した例はない。
【００１３】
【発明が解決しようとする課題】
食品製造工場の製品管理において、前記のような、抜き取り検査及び破壊計測の欠点（検査漏れ、製品ロス、タイムラグ）を改善することを目的に、分光分析法を利用して、内部に浸透性のある特定波長（４００ｎｍから５０，０００ｎｍの範囲）の光を用いて、ゲル形成性食品の中間製品又は包装された製品の品質、特に成分含量変化を伴わない物性変化について非破壊式又は非接触式全数検査システムを確立し、また光切断法と画像解析手法と併用してゲル形成性食品内部の二次元又は三次元位置における品質、均一性を評価することが本発明の課題である。
【００１４】
【課題を解決するための手段】
本発明者らは、実際に、同一成分でありながら物性がゾルからゲル（または液体から固体）に変化する食品、例えばタンパク質の変性や多糖類等の相互作用（一般に高分子の高次構造の変化や、高分子間のイオン結合、水素結合、疎水結合、ＳＳ結合等の形成によるゲル化）が起こる豆腐、卵豆腐、蒲鉾、チーズ等の品質判定や、豆乳や牛乳の粘性や凝固変質等の品質判定に可視光から近赤外線、赤外線の分光分析法の応用を検討した結果、特定波長領域の光の吸光度又はその演算値が、特定の品質を反映していることを見出し、本発明に至った。
【００１５】
またレーザー光線を照射光源とし、ＣＣＤカメラを受光装置として、製品コンベア上にそれらを設置した、いわゆる光切断法による三次元計測システムを構成し、特定波長領域の光を用いて、ゲル形成性食品内部の二次元又は三次元位置における品質を判定する、実用上安価な非接触・非破壊自動計測システムを実現し、本発明を完成させた。
【００１６】
【発明の実施の形態】
本発明請求項１記載の実施形態は、可視光から赤外線領域（４００〜５０，０００ｎｍ）、特に４００〜１，０００ｎｍの可視光から近赤外線領域の、単一又は複数の特定波長領域の光を用いる、分光分析計測システム方法であって、対象がゲル形成性食品の物性、特に成分含量変化を伴わない物性変化であることを特徴とする、ゲル形成性食品の品質判定方法である。
【００１７】
本発明で用いる分光分析計測システムは、図１（ａ）に示すように、光を被対象物１に投光する光源２と、その反射又は透過光５を受光し吸光度を測定する受光装置３を備え、必要に応じて、投光された光４又は受光された光５の特定波長領域を分離する分光装置６を付設するか、又は光路調整器７を付設するか、又はその吸光度を数学的前処理してあらかじめ作成された回帰式を用いて被対象物の特性関連値を出力する演算装置（コンピューター、シーケンサー等）を付設する、いわゆる分光分析計測システムである。図１（ｂ）は透過照射の場合の概略図である。なお、図示はしないが、側面方向でも同様に、透過反射照射や透過照射のシステムを適用できる。
【００１８】
特定波長領域は、製品種類や目的の品質項目毎に、その評価値と各波長領域の吸光度やその演算値との相関分析や重回帰分析、判別分析等の統計解析を行って選択される。４００〜５０，０００ｎｍまでの範囲には、可視光、近赤外線、（中）赤外線が含まれ、有機化合物やその官能基に由来する結合音やその倍音による吸収が存在する波長領域であるので、各種化合物に特有の吸収スペクトルが得られ、成分濃度との相関が得られやすい。特に４００〜１，０００ｎｍの、短波長の近赤外領域から可視光は、水による吸収が少なく、高水分（６０％ｗｔ以上）の食品には浸透性が高く、食品内部の情報を得る目的では好ましい。
【００１９】
可視光領域では色の三刺激値（例えばＲＧＢ表色法によるＲＧＢ値、ＣＩＥハンター表色法によるＬａｂ値等）等を用いてもよい。
【００２０】
近赤外線領域は、一般に波長域１（結合音域）１，８００〜２，５００ｎｍ、波長域２（１次倍音域）１，４００〜１，８００ｎｍ、波長域３（２次倍音域）７００〜１，１００ｎｍと３区分に細かく分けられる。特に８０％ｗｔ以上の高水分食品に適し、透過性の高い波長域３を用いることが好ましい、とされている（佐藤哲生、第５回非破壊計測シンポジウム講演要旨集ｐ８〜１４；岩元・魚住、日本食品工業学会第３２巻第９号ｐ６８５〜６９５）。
【００２１】
包装された製品の場合、包材の吸収波長を避けた波長領域を選択することによって、内部の食品の情報を得ることができる。また、包材の色や材質、印字の影響や、表面の付着水やパック水等の水の影響、食品の表面の焼き目や揚げ皮、着色等の影響も、同様にそれらの吸収波長領域を避けることで、的確に内部情報を得ることができる。
【００２２】
包材は特定波長領域の光を一部透過できるものであれば、基本的にどのような材質、厚みでも適用できる。肉眼では透けて見えなくても、強力な可視光線や赤外線を透過すれば、利用できる。ただし、アルミ蒸着された包材など、全く光を透過しない包材には適用できない。また、フィルム面以外に、容器側（底面や側面）からでも、投光又は受光が可能である。容器の形状は四角、丸形、チューブ状等、特に制約はない。
【００２３】
被対象物に照射する光源は、太陽光（自然光）や人工光源である蛍光灯、レーザー、ハロゲンランプ、白熱電球、単色ランプ、高圧放電ランプ（水銀灯、メタルハライドランプ）、キセノンランプ、発光ダイオード（ＬＥＤ）などである。また、フィルター、プリズム、回折格子等によって、特定の波長領域に制限した光を用いてもよい。また光源の点灯時間は高周波点灯等の常時点灯又はパルス点灯などが用いられる。被対象物の厚みがある場合や包材が光不透過性である場合等は高出力の光源を用いる。またリング状の光ファイバー灯や蛍光灯や白熱球や棒状蛍光灯など市販のあらゆる形態を用いることができる。また投光はスポット状、スリット状等の光路調整器や偏光フィルター、バンドパスフィルター等の光学フィルターによって制限するか、レンズによって集光、拡散させてもよい。受光装置に係わる投光照明は反射照明、透過照明、同軸照明、拡散照明等のいずれの形でも適用可能である。
【００２４】
光の照射を受けた被対象物からは、表面で跳ね返る反射光、被測定物を透過または散乱し貫通する透過光、照射光の一部が内部を透過・散乱して放射される透過反射光、色素などの励起によって起こる遅延光（蛍光）などがあり、少なくともその１つを利用する。また、その測定光もバンドパスフィルター等の光学フィルターや回折格子等の分光器によって、特定の波長領域に制限（分光）して用いてもよい。予め、投光の波長領域を制限する場合は、受光側の波長制限は不要である。
【００２５】
受光は光電管やダイオード素子により構成された受光装置、又は市販のＣＣＤカメラ、ＭＯＳ型カメラ等のエリアセンサー、リニアセンサー、赤外線カメラ、分光光度計、色彩色差計、果物糖度計、米食味計等を利用することによって、電気信号に変換される。あらかじめ標準白色板や比較対照試料で得ている基準値と相対的な比較によって、特定波長領域の相対的な吸光度として測定される。
【００２６】
被対象物は、静止状態でもよいが、工程上、コンベア搬送、シュートなどの搬送装置によって連続又は間欠移動してもよい。小規模であれば、樹脂コンテナ等に整列されて納められた状態でも構わない。また液体又はゾル状であればポンプ等によってパイプライン中を移動している場合やタンク中に滞留している場合も含まれる。
【００２７】
本発明の適用できるゲル形成性食品とは最終状態が固体又はゲル状である食品や、最終状態は液体又はゾル状であって、何らかの原因で、粘性増加や固体又はゲル状に変化する食品を指す。最終製品又は中間製品のどちらでも良い。形状も立方体、直方体、円柱、カップ状、球状など特に限定されない。包装の有無も限定しないが、少なくとも、特定波長領域の光が通過できる部分を有する包装材料で包装された場合に限られる。
【００２８】
本発明の適用できるゲル形成性食品のうち、最終状態がゲル状の食品は下記に示すような食品である。
【００２９】
タンパク質性ゲル形成性食品では、絹ごし豆腐・充填豆腐・木綿豆腐・寄せ豆腐（おぼろ豆腐）等の豆腐類、厚揚げ・生揚げ・薄揚げ・薄揚げ・寿司揚げ・がんもどき等の油揚生地、凍り豆腐やその凍結前後の生地、すし揚げ・厚揚げ・生揚げ・薄揚げ・すし揚げ・がんもどき等の油揚類、湯葉や湯葉豆腐、大豆タンパク質ゲル・豆乳ヨーグルト等の大豆加工食品、蒲鉾・竹輪・揚げ蒲鉾、魚肉ソーセージ等の水産練り製品、卵豆腐、ゆで卵、プリン、茶碗蒸し等の卵製品、チーズ・ヨーグルト等の牛乳加工製品、ゼラチンを用いたゲル状食品、うどん・素麺・ラーメン・パスタ・生麩・パン生地・ビスケット等の菓子生地等の小麦加工食品、蕎麦などである。
【００３０】
デンプン・多糖類性ゲル形成性食品ではコンニャクやコンニャクゼリー、ところてん、ういろう、煎餅やかきやま等の菓子生地、餅、胡麻豆腐、杏仁豆腐や豆花等のゲル化剤を使用したゼリー状食品である。ゲル化剤とはゼラチン、寒天、カードラン、カラギーナン、デンプン等の食品添加物である。
【００３１】
また、本発明に適用される液体又はゾル状の食品は、加工や保存中に凝固又はゲル化する性質を有する、例えば豆乳飲料、生卵、すり身、牛乳や牛乳加工飲料、寒天飲料等である。
【００３２】
本発明において対象となるゲル形成性食品の品質は個々に様々であるが、一般に豆腐や卵豆腐のようなゲル状食品の場合、一般に物性（硬さ・弾力・保水性等）、外観（色調、重量、光沢、きめ等）、味などが挙げられる。これらの品質は、同一成分でも、加熱、攪拌、時間などの微妙な加工条件によって影響が現れる。なお、形状（大きさ、欠損等）、タンパク質、脂肪、糖質、繊維質等の栄養成分、アミノ酸や核酸等の旨味成分、ビタミン類、一般細菌数や大腸菌群数、異物混入なども本発明の対象となる品質項目である。
豆乳飲料のような液状食品の場合も、上記以外に、物性（粘度・流動性・凝固や沈殿物等の有無）などの項目が挙げられる。
【００３３】
分光分析法の常套手段に従って、未知試料を測る前に、あらかじめ品質の異なる製品試料を複数測定し、回帰式を求めておく必要がある。先ず、品質の異なる良品から不良品までの製品に関して、一定条件において各波長領域の吸光度値を測定する。その製品品質（客観的・主観的に測定される評価値）と、各波長領域の吸光度又はその吸光度の多次微分値（例えば２次微分）やそれらの四則演算値、対数値、他の特定波長領域の吸光度との演算値等の数値と、その吸光度との相関分析、判別分析、重回帰分析などの統計解析又はニューロコンピューターの学習を行い、回帰式又はニューロコンピューター学習構造を求める。なお、本発明では特に製品の大きさ、形状、水分等の成分値の条件はほぼ固定条件にして、回帰式を用意した。なお、実用上は、各製品によって個別の回帰式を求める必要があり、面倒ではある。ニューロコンピューターの学習機能や認識を利用することによって、ユーザー側で現場レベルでのチューニングが容易になる。ニューロコンピューターを利用する利点は、非線形な因果関係に適していて、煩雑な統計計算が不要であり、複雑な自然現象に対して、比較的精度の高い判定を行うことも可能である。なお、選択された特定波長領域は、一般に特定の化学的構造（官能基）や特定の化合物に帰属されるが、学術的な説明ができなくても、実用上、因果関係が成立すれば問題ない。
【００３４】
外乱（例えば、特に温度や食品の大きさ等）による補正を行う必要がある場合、その外乱に影響の少ない特定波長領域の吸光度を選択し、その吸光度を差し引くこともある。ただし、加工食品の場合、形状や大きさや製品温度は、その製品の特定の工程では一定であるので、それぞれの製品によって判別式を決定しおけば、相対的な比較によって、判別が可能である。また吸光度の２次微分値等を用いれば、その外乱の影響を少なくできる。逆に、本発明によって、そのような温度変動や量目の変化も検知することも可能である。
【００３５】
図２は本発明請求項２記載のゲル形成性食品の品質判定装置についての概要図である。請求項１記載のゲル形成性食品の品質判定方法において、被対象物１が移動装置９によって直線的に、連続又は間欠移動する工程を設けて、移動方向に対して直角方向に直線状に投光される光４（レーザー光等）を用いて、被対象物に対して垂直方向より斜めに投光（入射角θ；−９０°＜θ＜９０°、図３（ａ）参照）し、その投光角度θより垂直方向、好ましくは垂直に受光するように配設（反射角η；θ＞０の時は０＜η≦９０°、θ＜０の時は９０°＜η＜１８０°、図３（ａ）参照）した受光装置３（ＣＣＤカメラ等のエリアセンサー）を備えた、いわゆる光切断法による三次元形状測定システムを応用する。
【００３６】
その直線状光線が被対象物表面と底面を横切る位置の間における、特定波長に限定された透過光又は反射光の吸光度の二次元分布を画像として取り込み、各画素の吸光度を深さにおける吸光度に換算することを、被対象物が移動する間に連続的に行い、得られた平面又は断面画像を合成し、被対象物の内部全体の二次元又は三次元位置における性状判定を行うものである。画像の取り込みは演算処理能力によるが、極短時間で繰り返し、各画像データから二次元又は三次元の各位置において積算、平均、深度による補正など演算処理を行い、二次元又は三次元の吸光度パターンを得る。または複数のスリット光を一度に照射し、一度に画像を取り込んでもよい。また移動方法は直線的移動、回転的移動等の一定した移動方法であれば特に限定しない。
入射角θや反射角ηは、包装された製品中や外表面に水が存在する場合、表面の全反射を避けるよう、両角度を異なる角度に配置にすることが望ましい。
【００３７】
特定波長領域の光は、１領域だけでなく、複数の領域の光を必要とする場合、分光器６を短時間に調節する（回折格子の角度を機械的に変更する、光学フィルターを回転式に切り替える、等）ことによって、数種の領域の吸光度を測定することが可能である。また、コスト的に若干高くなるが、分光器６や受光装置３を複数設けることによって、さらに高速処理が可能になる。なお、本発明では直線状の投光照明は被対象物の底面から行ってもよく、その透過光について同様な画像測定と判定を行うことができる。
【００３８】
モデル的には表面の線状反射光から距離Ｌだけずれた位置における線状反射光は被対象物中に散乱又は吸収されながら、一部の光は距離Ｌ÷ｃｏｓ（θ）だけ進んだ位置、即ち被対象物中の深さＬ×ｔａｎ（θ）の位置から、更に一部の光が垂直方向に向い、散乱又は吸収されながら距離Ｌ×ｔａｎ（η）だけ進んで、被対象物表面から放射され、透過反射光として受光装置に取り込まれる。見かけの光路長はＬ×（１÷ｃｏｓ（θ）＋ｔａｎ（η））
【００３９】
実際に、不透明ゲル形成性食品や乳濁液では、入射光の直進方向の周辺への散乱が大きく影響するので複雑ではあるが、光の強度の増加や被対象物の厚みを薄くすることや、水やタンパク質等の大きな吸収のない、透過しやすい特定波長を選択することによって、その影響を最小限にできる。なお、被対象物の厚みは包材や内部食品の光透過性にもよるが、２０ｃｍ以下、好ましくは５ｃｍ以下であれば十分適用できる。また透過性の高い波長領域は５００〜１，０００ｎｍの範囲である。
【００４０】
同一製品において、その入射から透過反射までの光路上、内部組織が均一であれば、どこの測定画像もその製品特有の同じ吸光度パターンを示すが、もし一部分に不均一な部分不具合があれば、二次元又は三次元吸光度パターンの一部分の吸光度が増減するので、不良品を判別できる。また木綿豆腐のように内部に「す」（水や空気が詰まった空隙）の状態やゲル形成性食品特有の気泡の抱き込み状態も、吸光度パターンに斑点状に強弱が表れるので判別できる。充填豆腐のパック内の離水状態も、水の吸収がある波長の吸光度パターンを撮れば、その離水状態を判別できる。これら吸光度パターンのそれぞれの位置について回帰式を用いて、三次元位置における硬さ（破断力）を求めることができる。またパイプライン中で豆乳を凝固させ、豆腐を連続的に製造するシステムにおいては、配管状に窓を設けて、本発明を適用した場合、パイプ中を移動する豆腐の断面における吸光度パターンをモニターでき、万一の凝固不良を検知することができる。なお、目的の品質が比較的単純な回帰式で表せる場合、単に、回帰式まで用いなくても、生産ライン上、相対的な吸光度パターンの変化をモニター（パターン認識）し、一般的な画像処理で閾値を越える、異常と疑われる製品があればシーケンサーに信号を出力し、搬送コンベア９から排出装置によって不良品を排除する。
【００４１】
エリアセンサーのうち、特にＣＣＤカメラは市販の安価なタイプで十分であり加工食品工場に導入しやすい。同時に二次元のデータを扱えるため、個々の被測定物の全体（全面）を評価でき、食品の状態が不均一である場合や、ムラのある場合にも適用しやすい。また、被対象物との距離をおいて非接触測定できるため、既存ラインに設置しやすく、機械動作機構を省いて高速で処理できる。データを２値化処理し、既存の画像解析技術を利用できる。それによって、均一性、異物混入、欠け、割れなどその他の情報を得ることができる。成分（タンパク質、脂質、水分等）の測定も合わせて行う場合、１，０００ｎｍ以上の近赤外線領域を用いることが好ましい。
【００４２】
【実施例】
ゲル形成性食品の一例として、豆腐について、図１（ａ）に示すような分光分析計測システムを用いた実施例を以下に述べる。
【００４３】
（実施例１）
大豆および浸漬条件は平成１２年度米国産ＩＯＭで１４時間、平成１２年度国産大豆タチナガハとエンレイ１：１混合したもので２０時間、井戸水流水下、１５℃で浸漬した。
【００４４】
生大豆８ｋｇ分の漬大豆を豆乳製造プラント（ＮＳ２０００Ｓ、高井製作所製）で５分１０３℃まで蒸気煮沸を行い、直ちに豆乳絞り機（シリウス１連、高井製作所製）にて、豆乳とオカラに分離した。得られた豆乳は約１３％ｂｒｉｘ（固形分濃度約１１％ｗｔ）であった。豆乳は１０℃になるまで急速に冷却し、所定量の凝固剤を混合し、ＰＰ製白色パック（２Ｂ）に充填し、ＰＰ製フィルムで溶着シールした。所定温度に合わせた湯浴中で、所定時間静置して充填豆腐を試作した。
【００４５】
上記と同様に分離直後の温豆乳（約８０℃）と乳化凝固剤（マグネスファインＴＧ、花王製）を使用して、ホット充填豆腐製造装置（高井製作所製）によって、ＰＰ製白色パック（京型）に充填し、ＰＰ製フィルムで溶着シールした。
【００４６】
凝固剤種類は塩化マグネシウム（ホワイトニガリ、ナイカイ塩業製）、硫酸カルシウム（すまし粉、ニットー製）、グルコノデルタラクトン（ＧＤＬ、藤沢薬品製）をそれぞれ豆乳重量に対して、０．０５、０．１、０．１５、０．２、０．２５、０．３、０．４、０．５％を添加した。また塩化マグネシウムとＧＤＬを１：１にブレンドした場合も同様に調製した。
【００４７】
加熱凝固条件は、未加熱、４０、５０、６０、７０、８０℃の湯浴温度にて、１０、６０分間加熱し、凝固した。
これらの条件の組み合わせによって、約１００個の豆腐を試作した。
【００４８】
それぞれの豆腐の品質測定は、レオメーター（不動工業製ＮＲＭ−２００２Ｊ）によって破断力等の測定（充填豆腐の包装フィルムを剥がした状態で、２３ｍｍφのプランジャーを６ｃｍ／分の速度で侵入させ、破断した時点の応力を測定）、パネラー数名による試食評価（外観、色、臭い、味、テクスチャーについて１０段階の採点し、合計を１００点満点として換算）、２時間放置前後の重量差から離水率の測定を行った。
【００４９】
分光光度測定機は市販のクボタ製フルーツセレクターを使用し、室温にて充填豆腐の包装フィルムの上面中央部に測光センサーを接触させて、波長５００〜１，００８ｎｍの間で２ｎｍ毎の吸光度を測定した。また、波長４００〜７００ｎｍの間で、約４５０ｎｍ、約５６０ｎｍ、約６１０ｎｍに分光感度を有するセンサーを内蔵した色彩色差計（ミノルタ製ＣＲ−３００、拡散照明垂直受光式）によってＬ値、ａ値、ｂ値（ハンター表色法）の測定も行った。
【００５０】
得られたデータについて市販の統計解析ソフト（マイクロソフト製ＥＸＣＥＬ２０００等）を用いて、相関分析、重回帰分析、分散分析を行い、重回帰式を求めた。
【００５１】
結果の一例として、塩化マグネシウム１００％充填豆腐の典型的な結果を表１、その吸光度スペクトルを図４、吸光度の２次微分スペクトルを図５に示した。図４から６８０〜９５０ｎｍにおける大きな吸収が存在することが分かった。特に８００〜８４０ｎｍ付近が特徴的で、最も大きな吸収を示している波長８２０ｎｍに関して、吸光度と破断力との関係を図６に示した。両者の間には高い負の相関（相関係数０．９９、ｎ＝６）が認められた。
【００５２】
【表１】

【００５３】
全データ（ｎ＝１００）について、重回帰分析を行った結果について表２に示した。破断力、離水率、試食評価共に吸光度の２次微分値との相関が高かった。
【００５４】
【表２】

【００５５】
塩化マグネシウム１００％の充填豆腐（ｎ＝２２）に限定した場合の結果を表３に示した。破断力ではＬａｂ値との相関が高く、図７に示すとおり、重相関係数０．９６３であった。離水率や試食評価も図８、図９に示すとおり、全データよりも高い相関を示した。
【００５６】
【表３】

【００５７】
マグネスファイン１００％のホット充填豆腐（ｎ＝２２）についても、破断力はＬａｂ値、吸光度やその２次微分値と高い相関（各重相関係数０．９７０、０．７１４、０．８４４）を示した。
【００５８】
以上からそれぞれの製品種類毎に作成した回帰式や判別式を用いることによって、柔らか過ぎ、離水が多過ぎ、試食評価が低い、というような不良品を明確に判別できることを確認した。
【００５９】
（実施例２）
図２に示した三次元測定システムを利用して、前記実施例１で得られた８２０ｎｍ吸光度を変数とした回帰式を用いて、充填豆腐内部の品質評価、特に破断力の予測を試みた。図３（ａ）に示したように均一な製品であれば、直線状のレーザー光の入射位置から入射光が底面に当たる位置の範囲で、ＣＣＤカメラによって画像を測定すると、光路長に応じた減衰した分布が得られる。しかし、図３（ｂ）に示したように外観上は正常であるが、内部に未凝固の部分（豆乳状）が存在する豆腐の場合、その未凝固部分を通過した光はより吸収され、その部分だけ暗い画像となるので、容易に不良品であることを判別できた。
【００６０】
【発明の効果】
本発明の請求項１記載の食品品質判別方法によれば、ゲル形成性食品において、その反射光又は透過光を受光し、特定波長領域（４００ｎｍから５０，０００ｎｍの範囲）の光について吸光度を測定し、ゲル状食品の品質を客観的に判定し、半凝固や未凝固などの不良品を判別することができる。また、一部分が光透過性であれば、その包材越しに中身の品質を評価することができる。その結果、ゲル形成性食品の自動製造ラインで、客観的な全数検査が容易になり、不良品を確実に除去でき、ロスやＰＬ問題等のトラブルを未然に防ぐことが可能になる。
【００６１】
本発明請求項２記載のゲル形成性食品の品質判別装置によれば、可視光領域から近赤外光領域に感度のある安価な市販ＣＣＤカメラ等のエリアセンサーで安価に構成し、遠隔測定（非破壊かつ非接触測定）し、汎用の画像解析処理を活用しながら、受光の特定波長領域の光を用いた光切断法を利用して、外観や成分だけでなく、ゲル形成性食品の内部品質（物性等）の二次元又は三次元位置に関する情報を得て、一層精度の高い迅速な製品の合否判定をすることができる。
【図面の簡単な説明】
【図１】（ａ）本発明請求項１および実施例１記載のゲル形成性食品の分光分析計測システム（透過反射型）の説明図
（ｂ）本発明請求項１および実施例１記載のゲル形成性食品の分光分析計測システム（透過型）の説明図
【図２】本発明請求項２記載のゲル形成性食品の三次元分光分析計測装置の説明図
【図３】（ａ）本発明実施例２記載の均一なゲル形成性食品の三次元分光分析計測の説明図
（ｂ）本発明実施例２記載の不均一なゲル形成性食品の三次元分光分析計測の説明図
【図４】本発明実施例１のニガリ１００％充填豆腐の透過反射光による吸光度スペクトルの説明図
【図５】本発明実施例１のニガリ１００％充填豆腐の透過反射光による吸光度２次スペクトルの説明図
【図６】本発明実施例１のニガリ１００％充填豆腐の透過反射光による吸光度と破断力の関係の説明図
【図７】本発明実施例１のＬａｂ値を用いた重回帰式による予測破断力と実測破断力との相関関係の説明図
【図８】本発明実施例１の吸光度２次微分値を用いた重回帰式による予測破断力と実測破断力との相関関係の説明図
【図９】本発明実施例１の吸光度を用いた重回帰式による予測破断力と実測破断力との相関関係の説明図
【符号の説明】
１被対象物
２光源
３受光装置
４照射光（投光）
５反射又は透過光
６分光器
７光路調整器
８スリット光用光路調整器
９搬送コンベア[0001]
BACKGROUND OF THE INVENTION
The present invention includes, for example, gelled foods such as tofu, egg tofu, rice cake, and cheese, and soy milk, raw eggs, surimi, and milk that have the property of solidifying or gelling during processing and storage. The present invention relates to a method for determining the quality inside a product in a non-destructive or contacted manner by spectroscopic measurement for a food-gel-forming food and an apparatus for determining quality at a two-dimensional or three-dimensional position.
[0002]
[Prior art]
In quality control at food manufacturing plants, in general, quality control personnel can make objective judgments on the quality of intermediate products or final products by using test equipment or chemical analysis methods for sampling inspections that are conducted at each production lot or periodically. There is a problem that it takes time until an inspection result is obtained and an inspection omission. In addition, during production, the operator pays attention to all products and judges the quality by visual inspection or touch, but carelessness and misjudgment cannot be eliminated.
[0003]
Gel-forming foods change from liquid or sol form to solid or gel form by heating, cooling, addition of coagulant, fermentation, enzyme reaction, etc. It is food to let you. As with general foods, a single product is unsealed by a product sampling test, and the gel strength is evaluated using a physical property tester such as a rheometer. Because it is a destructive test, that part is lost. Furthermore, if a defective product is detected, all the corresponding lots must be discarded even if the defective product is a part. Like gelation, processing operations and quality judgments that involve changes in physical properties greatly depend on the subjectivity, experience, and intuition of workers. Although it is easy to identify defective products before packaging in an intermediate state or intermediate product in a gel state, a defective product with unevenness or defects may be shipped for some reason on an automated line. In commercial products such as pudding and yogurt, a space is provided at the top of the cup to make it easy to visually determine the gelled state. However, there are problems such as oxidation by air, generation of microorganisms and loss of shape. Moreover, when it gelatinizes after packaging like filling tofu and egg tofu, it may be difficult to distinguish the quality of the gelation at first glance, and only a feeling such as touch may be relied upon. There is no abnormality in appearance, but if there is an abnormality inside, the defective product may be overlooked. In rare cases, incompletely solidified products are shipped, complaints from customers and consumers, product collection, and even liability issues arise, which puts a heavy burden on the food manufacturer.
[0004]
The quality of gel-forming foods generally includes quality items such as shape (size, defect, weight, etc.), appearance (color, weight, gloss, texture, etc.), texture (hardness, elasticity, water retention), and taste. Can be mentioned. Some are measured objectively using measuring equipment, but the taste and texture are currently judged by human taste.
[0005]
In addition, in the case of liquid or sol foods (drinks) such as soy milk and milk, the product is normally in a liquid or sol state, but for some reason, aggregates, precipitates / floats, increased viscosity, gelation, etc. It may be a defective product.
[0006]
For the purpose of inspection of the surface shape of the product, for example, a method of determining defects in packaged tofu by irradiation with slit light using a fluorescent lamp and image processing with a CCD camera (Japanese Patent Laid-Open No. 08-217029), spot projector And a method of determining by arranging a large number of light receiving sensors (Japanese Patent Laid-Open No. 10-62132) is disclosed. These methods simply measure the shape of the outer shape, but are not sufficient because they are affected by the color of the container and printing on the film.
[0007]
Regarding tofu before packaging, etc., a method (Japanese Patent Laid-Open No. 2001-133233) for determining a defect using a general light cutting method is disclosed, but it cannot be applied to a packaged product.
[0008]
There are disclosures (Registered No. 2931838 and Japanese Patent Application Laid-Open No. 2001-242100) of an image analysis apparatus using X-rays as a liquid medicine or food coagulation determination apparatus. These devices need to contain bubbles in the object, and the former is a method of measuring the state of dispersed bubbles by vibrating the object. Both can be applied to liquid foods, but not to gel foods. In addition, safety management is necessary to handle radiation.
[0009]
As a method for evaluating the component of the object using visible light spectroscopy, for example, the transmitted or reflected light of the object such as fruit is divided, passed through a plurality of specific wavelength filters, and the difference in absorbance is determined. Discloses a method for evaluating internal quality, for example, maturity (Japanese Patent Laid-Open No. 52-83274).
[0010]
An organic substance quality evaluation method (near infrared analysis 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 rice taste meter (Japanese Patent Publication No. 1-49890) for measuring the content of amylose and the like, a fruit sugar content meter (Japanese Patent No. 2517858, JP-A-1-301147) for measuring sugar and the like are put into practical use. In addition, a method for evaluating the pH (hydrogen ion content) of a cocoa-containing food (Japanese Patent Laid-Open No. 2001-17084), a method for measuring differences in ingredients before and after the processing of food, and evaluating a processing state (Japanese Patent Laid-Open No. 7-12722, JP-A-7-107925) is disclosed. However, there are no examples of applying spectroscopic analysis with visible light, near infrared rays, or infrared rays to tofu and other gel-forming foods.
[0011]
In addition, in fruits and vegetables, a system for predicting hardness (maturity) using absorption in the near-infrared region has been devised (JP-A-1-301147). However, ripening of fruits and vegetables generally involves many chemical components. With change. A method (JP-A-7-270312) for directly measuring cooked rice after rice cooking treatment and evaluating physical properties such as hardness and stickiness in addition to appearance and taste has been disclosed. Is different, with some component changes.
[0012]
Both methods are analytical techniques based on changes in component content such as amylose and hydrogen ion concentrations. Physical property values such as hardness and stickiness are evaluated based on an indirect correlation with component values. Therefore, it is difficult to distinguish the quality when the components are the same and do not change and the physical properties are different. Until now, near-infrared spectroscopy has been researched and put to practical use for the purpose of nondestructive measurement of the difference in the amount of chemical components in a measured object. However, no case has yet been disclosed in which the internal quality of a gel-forming food is determined by applying a spectroscopic analysis method using visible light, near infrared light, or the like. In addition, as a three-dimensional measurement method of an object, a method using a light cutting method or a light cutting method using light in a specific wavelength range is a well-known technique used to capture the outline of an object with an image, and is a surface defect of the object. It was used exclusively for defect detection, distortion detection and external shape measurement. There is no example in which the light cutting method is applied to property determination in a two-dimensional or three-dimensional region inside the object.
[0013]
[Problems to be solved by the invention]
In the product management of food manufacturing plants, the use of spectroscopic analysis to improve the above-mentioned shortcomings of sampling inspection and destructive measurement (inspection omission, product loss, time lag). Non-destructive or non-contact method for the quality of intermediate products or packaged products of gel-forming foods, especially for physical property changes without changing the content of ingredients, using light of a specific wavelength (ranging from 400 nm to 50,000 nm) It is an object of the present invention to establish an all-inspection system and to evaluate the quality and uniformity at a two-dimensional or three-dimensional position inside a gel-forming food in combination with a light cutting method and an image analysis method.
[0014]
[Means for Solving the Problems]
The inventors of the present invention actually have the same component but the physical properties change from sol to gel (or liquid to solid), such as protein denaturation and interaction with polysaccharides (generally higher order structure of polymer). Such as tofu, egg tofu, rice cake, cheese, etc. that undergo changes, gelation due to the formation of ionic bonds, hydrogen bonds, hydrophobic bonds, SS bonds between polymers, etc., and the viscosity and coagulation alteration of soy milk and milk As a result of investigating the application of spectroscopic methods from visible light to near-infrared and infrared in the quality determination of light, it has been found that the absorbance of light in a specific wavelength region or its calculated value reflects the specific quality. It came.
[0015]
In addition, a laser beam is used as an irradiation light source, a CCD camera is used as a light receiving device, and they are placed on a product conveyor to form a three-dimensional measurement system using a so-called light cutting method. A non-contact, non-destructive automatic measurement system that determines the quality at the two-dimensional or three-dimensional position of the device is realized, and the present invention has been completed.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
The embodiment according to claim 1 of the present invention is capable of emitting light in a single or plural specific wavelength regions from visible light to infrared region (400 to 50,000 nm), particularly from visible light to near infrared region of 400 to 1,000 nm. A spectroscopic measurement system method to be used, wherein the target is a gel-forming food Physical properties, In particular, the present invention relates to a method for judging the quality of a gel-forming food, characterized in that it is a change in physical properties not accompanied by a change in component content.
[0017]
As shown in FIG. 1A, a spectroscopic measurement system used in the present invention includes a light source 2 that projects light onto an object 1 and a light receiving device 3 that receives the reflected or transmitted light 5 and measures the absorbance. If necessary, a spectroscopic device 6 that separates a specific wavelength region of the projected light 4 or the received light 5 is attached, or an optical path adjuster 7 is attached, or its absorbance is calculated mathematically. This is a so-called spectroscopic measurement system equipped with a calculation device (computer, sequencer, etc.) that outputs a characteristic-related value of an object using a regression equation created in advance by target preprocessing. FIG.1 (b) is the schematic in the case of transmission irradiation. Although not shown in the figure, a transmission / reflection irradiation system or a transmission irradiation system can be applied in the same manner in the lateral direction.
[0018]
The specific wavelength region is selected for each product type and target quality item by performing statistical analysis such as correlation analysis, multiple regression analysis, discriminant analysis between the evaluation value and the absorbance of each wavelength region and the calculated value. The range from 400 to 50,000 nm includes visible light, near-infrared light, and (medium) infrared light, and is a wavelength region where there is absorption due to the combined sound derived from the organic compound and its functional group and its overtone, Absorption spectra peculiar to various compounds are obtained, and correlation with component concentrations is easily obtained. In particular, visible light from the near-infrared region of a short wavelength of 400 to 1,000 nm is less absorbed by water, has high permeability to foods with high moisture (60% wt or more), and obtains information inside the food. Then, it is preferable.
[0019]
In the visible light region, color tristimulus values (for example, RGB values by the RGB colorimetric method, Lab values by the CIE Hunter colorimetric method, etc.) may be used.
[0020]
The near-infrared region generally has a wavelength region 1 (combined sound region) 1,800 to 2,500 nm, a wavelength region 2 (primary overtone region) 1,400 to 1,800 nm, and a wavelength region 3 (second overtone region) 700 to 1. , 100 nm and can be divided into 3 sections. It is said that it is particularly suitable for high-moisture foods of 80% wt or more, and that it is preferable to use a highly transparent wavelength region 3 (Tetsuo Sato, 5th Non-Destructive Measurement Symposium Abstracts p8-14; Iwamoto and Uozumi , Japan Food Industry Association Vol. 32, No. 9, p685-695).
[0021]
In the case of a packaged product, information on the internal food can be obtained by selecting a wavelength region that avoids the absorption wavelength of the packaging material. In addition, the influence of the color and material of the packaging material, the printing, the influence of water such as surface adhering water and pack water, and the influence of baking, fried skin, coloring, etc. on the surface of the food are also affected. By avoiding, internal information can be obtained accurately.
[0022]
As long as the packaging material can partially transmit light in a specific wavelength region, basically any material and thickness can be applied. Even if it cannot be seen through with the naked eye, it can be used if it transmits strong visible light or infrared light. However, it is not applicable to packaging materials that do not transmit light at all, such as aluminum-deposited packaging materials. In addition to the film surface, light can be projected or received from the container side (bottom surface or side surface). The shape of the container is not particularly limited, such as a square shape, a round shape, and a tube shape.
[0023]
Light sources to irradiate objects are sunlight (natural light) and artificial light sources such as fluorescent lamps, lasers, halogen lamps, incandescent lamps, monochromatic lamps, high-pressure discharge lamps (mercury lamps, metal halide lamps), xenon lamps, light-emitting diodes (LEDs) ) Etc. Further, light limited to a specific wavelength region by a filter, a prism, a diffraction grating, or the like may be used. As the lighting time of the light source, constant lighting such as high frequency lighting or pulse lighting is used. A high output light source is used when the object is thick or the packaging material is light-impermeable. In addition, all commercially available forms such as a ring-shaped optical fiber lamp, a fluorescent lamp, an incandescent bulb, and a rod-shaped fluorescent lamp can be used. The light projection may be limited by an optical path adjuster such as a spot shape or a slit shape, or an optical filter such as a polarizing filter or a bandpass filter, or may be condensed and diffused by a lens. The floodlight illumination related to the light receiving device can be applied in any form such as reflection illumination, transmission illumination, coaxial illumination, and diffuse illumination.
[0024]
Reflected light that bounces off the surface from the object irradiated with light, transmitted light that penetrates or scatters through the object to be measured, and transmitted reflected light that is radiated through a part of the irradiated light transmitted or scattered inside There is delayed light (fluorescence) caused by excitation of a dye or the like, and at least one of them is used. The measurement light may also be used after being limited (spectroscopic) to a specific wavelength region by an optical filter such as a bandpass filter or a spectroscope such as a diffraction grating. When the wavelength range of light projection is limited in advance, wavelength limitation on the light receiving side is not necessary.
[0025]
For light reception, a light receiving device composed of a phototube or a diode element, or an area sensor such as a commercially available CCD camera or MOS camera, a linear sensor, an infrared camera, a spectrophotometer, a color difference meter, a fruit sugar meter, a rice taste meter, etc. By using it, it is converted into an electric signal. The relative absorbance in a specific wavelength region is measured by a relative comparison with a reference value obtained in advance with a standard white plate or a comparative control sample.
[0026]
The object may be in a stationary state, but may be continuously or intermittently moved by a conveying device such as a conveyor conveyance or a chute in the process. If it is a small scale, it may be arranged in a resin container or the like. In addition, in the case of liquid or sol form, the case where it is moved in the pipeline by a pump or the like or the case where it is staying in the tank is also included.
[0027]
The gel-forming food to which the present invention can be applied is a food whose final state is solid or gel, or a food whose final state is liquid or sol, and for some reason increases in viscosity or changes to solid or gel. Point to. Either a final product or an intermediate product may be used. The shape is not particularly limited, such as a cube, a rectangular parallelepiped, a cylinder, a cup, or a sphere. Although the presence or absence of packaging is not limited, it is limited to at least a case of packaging with a packaging material having a portion through which light in a specific wavelength region can pass.
[0028]
Among the gel-forming foods to which the present invention can be applied, the foods in the final state of gel are foods as shown below.
[0029]
In protein gel-forming foods, tofu such as silken tofu, filled tofu, cotton tofu, jelly tofu, deep-fried, fresh-fried, thin-fried, thin-fried, deep-fried sushi, fried cancer, etc. Frozen tofu and dough before and after freezing, deep-fried sushi, deep-fried, freshly-fried, lightly-fried, sushi-fried, squirted, etc., processed soybean foods such as yuba, yuba tofu, soy protein gel, soy milk yogurt, strawberries, bamboo rings・ Fried crab, marine products such as fish sausages, egg tofu, boiled eggs, egg products such as pudding, steamed rice bowl, processed milk products such as cheese and yogurt, gelatinous foods using gelatin, udon, raw noodles, ramen, pasta, Wheat processed foods such as ginger, bread dough, confectionery such as biscuits, and soba.
[0030]
Starch / polysaccharide gel-forming foods are jelly-like foods that use konjac and konjac jelly, confectionery such as Toroten, Uiro, rice crackers and kakiyama, and gelling agents such as koji, sesame tofu, apricot tofu and bean flowers. . Gelling agents are food additives such as gelatin, agar, curdlan, carrageenan and starch.
[0031]
The liquid or sol food applied to the present invention has a property of solidifying or gelling during processing or storage, for example, soy milk drink, raw egg, surimi, milk or milk processed drink, agar drink, etc. .
[0032]
The quality of gel-forming foods that are the subject of the present invention varies individually. Generally, gel-like foods such as tofu and egg tofu generally have physical properties (hardness, elasticity, water retention, etc.) and appearance (color tone). , Weight, gloss, texture, etc.) and taste. These qualities are affected by subtle processing conditions such as heating, stirring and time even for the same components. In addition, the present invention includes shapes (sizes, defects, etc.), nutritional components such as proteins, fats, sugars and fibers, umami components such as amino acids and nucleic acids, vitamins, general bacterial counts, coliform counts, foreign matter contamination, etc. It is a quality item subject to
In the case of liquid foods such as soy milk beverages, in addition to the above, items such as physical properties (viscosity / fluidity / presence / absence of coagulation, precipitates, etc.) can be mentioned.
[0033]
Before measuring an unknown sample in accordance with conventional means of spectroscopic analysis, it is necessary to measure a plurality of product samples having different qualities in advance and obtain a regression equation. First, with respect to products from non-defective products with different qualities to defective products, the absorbance value in each wavelength region is measured under a certain condition. Product quality (evaluation value measured objectively and subjectively), absorbance in each wavelength region or multi-order differential value (for example, second derivative) of the absorbance, their four arithmetic operations, logarithmic value, other identification Statistical analysis such as correlation analysis, discriminant analysis, multiple regression analysis, etc. with a numerical value such as a calculated value of the absorbance in the wavelength region and the absorbance, or neurocomputer learning is performed to obtain a regression equation or a neurocomputer learning structure. In the present invention, the regression equation was prepared with the conditions of the component values such as the size, shape and moisture of the product being substantially fixed. In practice, it is necessary to obtain an individual regression equation for each product, which is troublesome. By using the learning function and recognition of the neurocomputer, tuning at the field level becomes easy on the user side. The advantage of using a neurocomputer is suitable for non-linear causal relationships, does not require complicated statistical calculations, and can make relatively high-precision judgments for complex natural phenomena. The selected specific wavelength region is generally attributed to a specific chemical structure (functional group) or a specific compound. However, even if scientific explanation is not possible, there is a problem if a causal relationship is established in practice. Absent.
[0034]
When it is necessary to perform correction due to disturbance (for example, temperature, food size, etc.), the absorbance in a specific wavelength region that hardly affects the disturbance may be selected and the absorbance may be subtracted. However, in the case of processed foods, the shape, size, and product temperature are constant in the specific process of the product, so if a discriminant is determined for each product, it can be discriminated by relative comparison. . If the second derivative value of absorbance or the like is used, the influence of the disturbance can be reduced. On the contrary, according to the present invention, it is also possible to detect such a temperature change and a change in quantity.
[0035]
FIG. 2 is a schematic diagram of the quality determination device for gel-forming food according to claim 2 of the present invention. The method for judging the quality of a gel-forming food according to claim 1, wherein a step is provided in which the object 1 is linearly moved continuously or intermittently by the moving device 9 and is linearly projected in a direction perpendicular to the moving direction. Using the emitted light 4 (laser light or the like), light is projected obliquely from the vertical direction with respect to the object (incident angle θ; −90 ° <θ <90 °, see FIG. 3A), It is arranged so as to receive light vertically from the projection angle θ, preferably vertically (reflection angle η; 0 <η ≦ 90 ° when θ> 0, 90 ° <η <180 ° when θ <0. A three-dimensional shape measuring system using a so-called light cutting method, which includes the light receiving device 3 (an area sensor such as a CCD camera) shown in FIG.
[0036]
A two-dimensional distribution of the absorbance of the transmitted or reflected light limited to a specific wavelength between the positions where the linear ray crosses the surface of the object and the bottom is captured as an image, and the absorbance of each pixel is converted to the absorbance at the depth. Conversion is performed continuously while the object is moving, and the obtained plane or cross-sectional image is synthesized, and the property is determined at the two-dimensional or three-dimensional position of the entire interior of the object. . Image acquisition depends on the calculation processing capability, but it is repeated in a very short time, and calculation processing such as integration, averaging, and depth correction is performed at each two-dimensional or three-dimensional position from each image data to obtain a two-dimensional or three-dimensional absorbance pattern. Get. Alternatively, a plurality of slit lights may be irradiated at a time to capture an image at a time. The moving method is not particularly limited as long as it is a constant moving method such as linear movement or rotational movement.
The incident angle θ and the reflection angle η are preferably arranged at different angles so as to avoid total reflection of the surface when water is present in the packaged product or on the outer surface.
[0037]
If the light of a specific wavelength region requires light of not only one region but also a plurality of regions, the spectroscope 6 is adjusted in a short time (mechanically changing the angle of the diffraction grating, rotating the optical filter It is possible to measure the absorbance of several regions. Further, although the cost is slightly increased, by providing a plurality of spectroscopes 6 and light receiving devices 3, higher speed processing can be performed. In the present invention, the linear floodlighting may be performed from the bottom surface of the object, and the same image measurement and determination can be performed for the transmitted light.
[0038]
As a model, the linear reflected light at a position shifted by a distance L from the linear reflected light on the surface is scattered or absorbed in the object, and a part of the light has advanced by a distance L ÷ cos (θ). That is, from the position of the depth L × tan (θ) in the object, a part of the light is further directed in the vertical direction and traveled by the distance L × tan (η) while being scattered or absorbed, and the object surface And is taken into the light receiving device as transmitted reflected light. Apparent optical path length is L × (1 ÷ cos (θ) + tan (η))
[0039]
Actually, opaque gel-forming foods and emulsions are complicated because the scattering of incident light to the periphery in the straight direction is greatly affected, but increasing the light intensity and reducing the thickness of the object By selecting a specific wavelength that does not absorb a large amount of water or protein and is easy to transmit, the influence can be minimized. In addition, although the thickness of a target object is based also on the light transmittance of a packaging material or an internal foodstuff, if it is 20 cm or less, Preferably it is 5 cm or less, and it can apply sufficiently. The wavelength region with high transparency is in the range of 500 to 1,000 nm.
[0040]
In the same product, if the internal tissue is uniform on the optical path from the incident to the transmission reflection, every measurement image shows the same absorbance pattern peculiar to the product, but if there is a non-uniform partial defect in a part, Since the absorbance of a part of the two-dimensional or three-dimensional absorbance pattern increases or decreases, a defective product can be identified. In addition, the state of “so” (voids clogged with water and air) as in the case of cotton tofu and the embedding state of air bubbles peculiar to gel-forming foods can be discriminated because they appear as spots in the absorbance pattern. The water separation state in the packed tofu pack can also be determined by taking an absorbance pattern at a wavelength where water is absorbed. The hardness (breaking force) at the three-dimensional position can be obtained by using a regression equation for each position of the absorbance pattern. In addition, in a system that coagulates soymilk in a pipeline and continuously produces tofu, when the present invention is applied with a pipe-like window, the absorbance pattern in the cross-section of tofu moving through the pipe can be monitored. In the unlikely event of coagulation failure, it can be detected. If the target quality can be expressed by a relatively simple regression equation, the change in relative absorbance pattern on the production line can be monitored (pattern recognition) without using the regression equation. If there is a product suspected of being abnormal in which the threshold is exceeded, a signal is output to the sequencer, and the defective product is removed from the conveyor 9 by the discharge device.
[0041]
Of the area sensors, in particular, a commercially available inexpensive camera is sufficient, and it is easy to introduce it into a processed food factory. Since two-dimensional data can be handled at the same time, the whole (entire surface) of each object to be measured can be evaluated, and it is easy to apply even when the state of food is uneven or uneven. In addition, since non-contact measurement can be performed at a distance from the object, it can be easily installed on an existing line, and can be processed at high speed without a mechanical operation mechanism. Data can be binarized and existing image analysis techniques can be used. Thereby, other information such as uniformity, contamination, chipping and cracking can be obtained. When measurement of components (protein, lipid, moisture, etc.) is also performed, it is preferable to use a near infrared region of 1,000 nm or more.
[0042]
【Example】
As an example of the gel-forming food, an example using a spectroscopic measurement system as shown in FIG.
[0043]
Example 1
Soybeans and soaking conditions were as follows: US-made IOM in 2000 for 14 hours, 2000 domestic soybean Tachinagaha and Enray 1: 1 mixed, and soaked at 15 ° C. under running well water for 20 hours.
[0044]
Steamed boiled soybeans for 8 kg of raw soybeans at a soymilk production plant (NS2000S, manufactured by Takai Seisakusho) for 5 minutes to 103 ° C, and immediately separated into soymilk and okara using a soymilk squeezer (Sirius 1 series, manufactured by Takai Seisakusho). did. The obtained soymilk was about 13% brix (solid content concentration about 11% wt). The soy milk was rapidly cooled to 10 ° C., a predetermined amount of a coagulant was mixed, filled into a PP white pack (2B), and sealed with a PP film. Filled tofu was made as a prototype by allowing it to stand for a predetermined time in a hot water bath adjusted to a predetermined temperature.
[0045]
A white pack made of PP (Kyoto) using hot-filled tofu production equipment (manufactured by Takai Seisakusho) using warm soy milk (about 80 ° C.) immediately after separation and an emulsifying coagulant (Magnesfine TG, manufactured by Kao Corporation) as described above. And welded and sealed with a PP film.
[0046]
The type of coagulant is magnesium chloride (white bittern, manufactured by Naikai Shigyo Co., Ltd.), calcium sulfate (summer powder, manufactured by Nitto), glucono delta lactone (GDL, manufactured by Fujisawa Pharmaceutical Co., Ltd.) 0.1, 0.15, 0.2, 0.25, 0.3, 0.4, 0.5% were added. Also, when magnesium chloride and GDL were blended 1: 1, the same preparation was performed.
[0047]
The heating and solidification conditions were as follows: unheated, heated at 40, 50, 60, 70, and 80 ° C. bath temperature for 10 to 60 minutes to solidify.
About 100 pieces of tofu were made as a trial by combining these conditions.
[0048]
The quality of each tofu was measured with a rheometer (NRM-2002J, manufactured by Fudo Kogyo Co., Ltd.) and the like (with a 23 mmφ plunger invaded at a rate of 6 cm / min with the packaging film of the tofu peeled off) Measure the stress at the time of rupture), taste evaluation by several panelists (score in 10 levels for appearance, color, odor, taste and texture, and convert the total to 100 points) The rate was measured.
[0049]
The spectrophotometer uses a commercially available KUBOTA fruit selector, and a photometric sensor is brought into contact with the center of the top surface of the packed tofu packaging film at room temperature to measure absorbance every 2 nm between wavelengths of 500 to 1,008 nm. did. In addition, the color difference meter (Minolta CR-300, diffuse illumination vertical light receiving type) with a built-in sensor having spectral sensitivity at wavelengths of about 450 nm, about 560 nm, and about 610 nm between wavelengths of 400 to 700 nm, L value, a value, The b value (hunter color method) was also measured.
[0050]
The obtained data was subjected to correlation analysis, multiple regression analysis, and analysis of variance using commercially available statistical analysis software (such as Microsoft Excel CEL2000) to obtain a multiple regression equation.
[0051]
As an example of the results, typical results of 100% magnesium chloride-filled tofu are shown in Table 1, the absorbance spectrum thereof is shown in FIG. 4, and the second derivative spectrum of absorbance is shown in FIG. It was found from FIG. 4 that there is a large absorption at 680 to 950 nm. FIG. 6 shows the relationship between the absorbance and the breaking force at a wavelength of 820 nm, which is particularly characteristic around 800 to 840 nm and exhibits the largest absorption. A high negative correlation (correlation coefficient 0.99, n = 6) was observed between them.
[0052]
[Table 1]

[0053]
The results of multiple regression analysis for all data (n = 100) are shown in Table 2. The breaking force, water separation rate, and taste evaluation were all highly correlated with the second derivative of absorbance.
[0054]
[Table 2]

[0055]
Table 3 shows the results when limited to filled tofu (n = 22) with 100% magnesium chloride. The breaking force had a high correlation with the Lab value, and the multiple correlation coefficient was 0.963 as shown in FIG. As shown in FIGS. 8 and 9, the water separation rate and the taste evaluation also showed a higher correlation than all the data.
[0056]
[Table 3]

[0057]
For magnes fine 100% hot-filled tofu (n = 22), the breaking force is highly correlated with the Lab value, absorbance, and its second derivative value (multiple correlation coefficients 0.970, 0.714, 0.844). showed that.
[0058]
From the above, it was confirmed that defective products such as too soft, too much water separation, and low tasting evaluation could be clearly identified by using regression formulas and discriminants created for each product type.
[0059]
(Example 2)
Using the three-dimensional measurement system shown in FIG. 2, the regression equation using the absorbance at 820 nm obtained in Example 1 as a variable was used to evaluate the quality inside the filled tofu, particularly to predict the breaking force. In the case of a uniform product as shown in FIG. 3A, when an image is measured with a CCD camera in the range from the incident position of the linear laser beam to the position where the incident light hits the bottom surface, the attenuation according to the optical path length Distribution is obtained. However, as shown in FIG. 3 (b), the appearance is normal, but in the case of tofu with an uncoagulated part (soy milk) inside, the light that has passed through the uncoagulated part is more absorbed, Since only that portion is a dark image, it was easily determined that the product was defective.
[0060]
【The invention's effect】
According to the food quality determination method of claim 1 of the present invention, in gel-forming food, the reflected light or transmitted light is received, and the absorbance is measured for light in a specific wavelength region (range from 400 nm to 50,000 nm). In addition, the quality of the gel food can be objectively determined, and defective products such as semi-solidified and non-solidified can be determined. Moreover, if a part is light-transmitting, the quality of the contents can be evaluated through the packaging material. As a result, an objective complete inspection can be easily performed on an automatic production line for gel-forming foods, defective products can be reliably removed, and troubles such as loss and PL problems can be prevented.
[0061]
According to the quality discrimination device for gel-forming food according to claim 2 of the present invention, it is inexpensively configured with an area sensor such as an inexpensive commercial CCD camera sensitive from the visible light region to the near-infrared light region. Non-destructive and non-contact measurement), utilizing general-purpose image analysis processing, and using a light-cutting method using light in a specific wavelength region of received light, not only the appearance and ingredients, but also the interior of gel-forming foods Information on the two-dimensional or three-dimensional position of quality (physical properties, etc.) can be obtained, and the pass / fail judgment of a product with higher accuracy can be made quickly.
[Brief description of the drawings]
FIG. 1A is an explanatory diagram of a spectroscopic analysis measurement system (transmission reflection type) for gel-forming food according to claim 1 and Example 1 of the present invention.
(B) Explanatory drawing of the spectroscopic analysis measurement system (transmission type) of the gel-forming food according to claim 1 and Example 1 of the present invention
FIG. 2 is an explanatory diagram of a three-dimensional spectroscopic analysis measuring apparatus for gel-forming food according to claim 2 of the present invention.
FIG. 3A is an explanatory diagram of three-dimensional spectroscopic analysis measurement of a uniform gel-forming food according to Example 2 of the present invention.
(B) Explanatory drawing of the three-dimensional spectroscopic measurement of the non-uniform gel-forming food according to Example 2 of the present invention
FIG. 4 is an explanatory diagram of an absorbance spectrum by transmitted / reflected light of 100% bittern tofu of Example 1 of the present invention.
FIG. 5 is an explanatory diagram of the absorbance secondary spectrum of transmitted / reflected light of 100% bittern tofu of Example 1 of the present invention.
FIG. 6 is an explanatory diagram of the relationship between absorbance and breaking force of transmitted and reflected light of 100% bitter tofu filled with Example 1 of the present invention.
FIG. 7 is an explanatory diagram of a correlation between a predicted breaking force and a measured breaking force based on a multiple regression equation using the Lab value of Example 1 of the present invention.
FIG. 8 is an explanatory diagram of the correlation between the predicted breaking force and the measured breaking force based on the multiple regression equation using the second-order absorbance value of the absorbance according to Example 1 of the present invention.
FIG. 9 is an explanatory diagram of a correlation between a predicted breaking force and a measured breaking force by a multiple regression equation using the absorbance of Example 1 of the present invention.
[Explanation of symbols]
1 Object
2 Light source
3 Light receiver
4 Irradiation light (projection)
5 Reflected or transmitted light
6 Spectrometer
7 Light path adjuster
8 Optical path adjuster for slit light
9 Conveyor

Claims

A quality determination method characterized by measuring a change in physical properties of a gel-forming food without a change in component content by a spectroscopic measurement system using light in a specific wavelength region from visible light to an infrared region (400 to 50,000 nm). .

The method for judging the quality of a gel-forming food according to claim 1, further comprising the step of moving the object by a moving device, and using light projected linearly in a direction perpendicular to the moving direction. A three-dimensional shape measurement system using a so-called light cutting method, comprising a light receiving device arranged to project light obliquely from a vertical direction with respect to an object and to receive light in a vertical direction from the light projection angle, The two-dimensional distribution of the absorbance of the transmitted or reflected light limited to a specific wavelength between the positions where the light beam crosses the surface of the object and the bottom is captured as an image and converted to the absorbance in the depth direction of the object This is performed continuously while the object is moving, and the quality determination device for the gel-forming food is characterized in that the quality of the object is measured at the two-dimensional or three-dimensional position based on the result.

By a spectroscopic measurement system using light in a specific wavelength region from visible light to infrared region (400 to 50,000 nm), breaking strength, hardness, elasticity, water retention, viscosity, fluidity, coagulation or precipitation of gel-forming food A quality determination method comprising measuring at least one of the presence or absence of an object and determining the quality of a gel-forming food.