郭志明,陳全勝,張 彬,王慶艷,歐陽(yáng)琴,趙杰文
?
果蔬品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)設(shè)計(jì)與試驗(yàn)
郭志明1,2,陳全勝1※,張 彬1,王慶艷2,歐陽(yáng)琴1,趙杰文1
(1. 江蘇大學(xué)食品與生物工程學(xué)院,鎮(zhèn)江 212013;2. 國(guó)家農(nóng)業(yè)智能裝備工程技術(shù)研究中心,北京 100097)
為滿足果蔬加工過(guò)程快速檢測(cè)和質(zhì)量控制的實(shí)際需求,研發(fā)近紅外光譜技術(shù)的低成本、實(shí)用化、小型化的果蔬品質(zhì)手持式檢測(cè)系統(tǒng)。在分析當(dāng)前近紅外光譜實(shí)用化過(guò)程的瓶頸問(wèn)題的基礎(chǔ)上,提出果蔬品質(zhì)的手持式檢測(cè)系統(tǒng)設(shè)計(jì)方案,闡述了硬件系統(tǒng)選擇和軟件系統(tǒng)構(gòu)建,介紹了檢測(cè)系統(tǒng)的工作原理;選用近紅外微機(jī)電系統(tǒng)的數(shù)字微鏡器件作為分光元件,以單點(diǎn)探測(cè)器獲取檢測(cè)信息,從而實(shí)現(xiàn)光譜檢測(cè)系統(tǒng)的微型化設(shè)計(jì)和系統(tǒng)成本的顯著降低。以檢測(cè)番茄為例,利用設(shè)計(jì)的手持式檢測(cè)系統(tǒng),獲取番茄900~1 700 nm范圍的近紅外光譜,利用先選擇特征波段再優(yōu)選波長(zhǎng)的建模策略,分別建立了番茄中番茄紅素和可溶性固形物含量的定量檢測(cè)模型;可溶性固形物含量模型的預(yù)測(cè)相關(guān)系數(shù)和預(yù)測(cè)均方根誤差分別為0.899和0.133%;番茄紅素模型的預(yù)測(cè)相關(guān)系數(shù)和預(yù)測(cè)均方根誤差分別為0.886和2.508 mg/kg。研究表明該系統(tǒng)能夠滿足果蔬品質(zhì)的快速無(wú)損檢測(cè)要求,可為實(shí)用化、小型化的手持式光譜檢測(cè)儀設(shè)計(jì)和開(kāi)發(fā)提供參考。
無(wú)損檢測(cè);光譜分析;農(nóng)產(chǎn)品;近紅外光譜;手持式檢測(cè)系統(tǒng);數(shù)字微鏡器件;果蔬品質(zhì)
果蔬產(chǎn)業(yè)是國(guó)民經(jīng)濟(jì)的基礎(chǔ)性、戰(zhàn)略性支柱產(chǎn)業(yè),在調(diào)整農(nóng)業(yè)產(chǎn)業(yè)結(jié)構(gòu)、延伸農(nóng)產(chǎn)品產(chǎn)業(yè)鏈、增加農(nóng)民收入等方面發(fā)揮了重要作用[1-2]。中國(guó)已穩(wěn)居世界水果和蔬菜第一大生產(chǎn)國(guó)和消費(fèi)國(guó),果蔬作為民生必需品受到人民群眾的極大關(guān)注和政府部門的高度重視[3]。傳統(tǒng)的化學(xué)方法費(fèi)時(shí)、費(fèi)力、檢測(cè)成本高且使用化學(xué)試劑,造成資源的嚴(yán)重浪費(fèi),無(wú)法滿足果蔬生產(chǎn)、流通、控制等環(huán)節(jié)快速檢測(cè)的需要,已成為制約果蔬業(yè)健康快速發(fā)展的瓶頸之一。為保證果蔬品質(zhì),提高國(guó)際競(jìng)爭(zhēng)力,果蔬品質(zhì)快速無(wú)損檢測(cè)技術(shù)與設(shè)備亟待開(kāi)發(fā)。
農(nóng)產(chǎn)品光電檢測(cè)技術(shù)具有高通量、多指標(biāo)同時(shí)檢測(cè)的優(yōu)勢(shì)[4-7]。其中近紅外光譜分析技術(shù)作為一種綠色分析技術(shù),具有快速、無(wú)損傷、可在線的優(yōu)點(diǎn),成為果蔬品質(zhì)快速無(wú)損檢測(cè)的首選技術(shù)[8-12]。目前的果蔬品質(zhì)近紅外無(wú)損檢測(cè)儀體積大、質(zhì)量大、攜帶不便[13-16],同時(shí)價(jià)格昂貴,如使用光譜儀一般采用線陣探測(cè)器,價(jià)格高[17-19],無(wú)法在食品、農(nóng)產(chǎn)品加工檢測(cè)行業(yè)推廣應(yīng)用。因此,果蔬品質(zhì)小型化、低成本、實(shí)用化的檢測(cè)系統(tǒng)具有廣闊的應(yīng)用前景[20-22]。另外,作為實(shí)用化的小型檢測(cè)設(shè)備必須具有低功耗的特點(diǎn),才能滿足原位、離線等多種工況的操作使用要求。
針對(duì)果蔬物料組織結(jié)構(gòu)和光傳輸特性,優(yōu)選近紅外光譜模塊,進(jìn)行光路結(jié)構(gòu)設(shè)計(jì)和軟硬件開(kāi)發(fā),研究果蔬主要品質(zhì)指標(biāo)快速、高精度、無(wú)損傷、智能化的檢測(cè)與評(píng)價(jià)方法,研制近紅外光譜技術(shù)的低成本、實(shí)用化、小型化的果蔬品質(zhì)手持式檢測(cè)系統(tǒng),對(duì)于保障果蔬品質(zhì)、實(shí)現(xiàn)果蔬加工增值、增強(qiáng)中國(guó)果蔬的國(guó)際競(jìng)爭(zhēng)力具有重要意義。
果蔬品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)硬件主要包括微型近紅外光譜儀、微型光源、控制與通訊模塊、移動(dòng)電源、橡膠墊圈和殼體,如圖1所示。本研究設(shè)計(jì)的手持式近紅外光譜檢測(cè)系統(tǒng)主要用于球形和類球形果蔬的品質(zhì)檢測(cè)。果蔬質(zhì)地軟易損傷,檢測(cè)時(shí)系統(tǒng)通過(guò)前置的軟橡膠墊圈接觸果蔬,同時(shí)將果蔬與光譜儀間的空間封閉,避免雜散光的干擾。工作時(shí)光源以特定角度照射到果蔬表面,光經(jīng)果蔬內(nèi)部傳輸后由光譜儀獲取體反射光,光信號(hào)轉(zhuǎn)化為電信號(hào)后由控制電路板傳輸?shù)秸粕想娔X(personal digital assistant, PDA)進(jìn)行顯示和存儲(chǔ)。檢測(cè)時(shí)通過(guò)觸控開(kāi)關(guān),打開(kāi)光源的同時(shí)觸發(fā)光譜儀采集果蔬的漫反射光譜,用于檢測(cè)果蔬的品質(zhì)。
1.1 微型近紅外光譜模塊
近紅外光譜是分子振動(dòng)的合頻和倍頻吸收,可以獲得待測(cè)對(duì)象的物理結(jié)構(gòu)和化學(xué)組分信息,在果蔬品質(zhì)檢測(cè)方面具有廣闊的應(yīng)用潛力[23-24]。現(xiàn)有光譜儀如光柵掃描式、傅里葉變換式、聲光可調(diào)濾光器式,雖然檢測(cè)精度較高,但儀器成本高、體積大,限制了實(shí)際生產(chǎn)中的應(yīng)用。隨著新型光譜儀微機(jī)電系統(tǒng)(micro electro- mechanical system, MEMS)的研發(fā)逐漸成熟,發(fā)展了基于MEMS技術(shù)的集成度高、體小結(jié)構(gòu)堅(jiān)固的新型便攜式近紅外光譜儀,MEMS技術(shù)利于大規(guī)模生產(chǎn)且降低儀器的成本。當(dāng)前小型近紅外光譜儀一般采用固定光柵結(jié)合線陣探測(cè)器的方式獲取近紅外信號(hào),價(jià)格有一定程度的降低但仍無(wú)法滿足低成本、實(shí)用化的實(shí)際需求,且存在信噪比低抗干擾能力弱的問(wèn)題。
數(shù)字微鏡器件(digital micromirror device,DMD)是集光處理與MEMS于一體的器件,因具有高分辨率、高亮度、高對(duì)比度、高可靠性、數(shù)字控制和響應(yīng)時(shí)間短等優(yōu)點(diǎn),廣泛應(yīng)用于數(shù)字光處理(digital light procession,DLP)系統(tǒng)中[25-26]。DMD由幾十萬(wàn)或數(shù)百萬(wàn)個(gè)微型數(shù)字可編程鏡片組成,微鏡固定在軛上,通過(guò)電極對(duì)微鏡產(chǎn)生靜電吸引,實(shí)現(xiàn)微鏡的轉(zhuǎn)動(dòng),進(jìn)而控制每個(gè)微鏡來(lái)產(chǎn)生特定模式的光信號(hào)獲取。因DMD是數(shù)字可編程的,可以根據(jù)用戶需要設(shè)置光譜分辨率和波長(zhǎng)范圍,調(diào)整積分時(shí)間,均衡光通量,可將信噪比提高到30 000:1以上,獲得比傳統(tǒng)光譜儀更快、更精準(zhǔn)的結(jié)果。
在農(nóng)業(yè)和食品工業(yè)中使用的近紅外光譜儀一般選用硅探測(cè)器,因?yàn)榫捎镁€陣探測(cè)器時(shí),配置硅探測(cè)器的近紅外光譜儀的價(jià)格為配置銦鎵砷探測(cè)器價(jià)格的1/5左右。本文選用DMD分光器件配置單點(diǎn)銦鎵砷探測(cè)器研發(fā)的近紅外光譜儀,成本僅為配置線陣硅探測(cè)器的近紅外光譜儀的1/3左右。采用DMD設(shè)計(jì)近紅外光譜儀,配置單點(diǎn)探測(cè)器,可避免使用高昂的線性探測(cè)器陣列的同時(shí)完成高性能的光譜儀。另外,DMD近紅外模塊具有微型化的特點(diǎn)。因此,選用DMD技術(shù)研發(fā)低成本、實(shí)用化、小型化、高精度的果蔬品質(zhì)手持式檢測(cè)系統(tǒng)具有技術(shù)優(yōu)勢(shì)和價(jià)格優(yōu)勢(shì)。
短波近紅外(700~1 100 nm)是分子振動(dòng)光譜三級(jí)和四級(jí)或更高級(jí)倍頻的吸收,光穿透能力較強(qiáng),但吸收較弱;長(zhǎng)波近紅外(1 000~2 500 nm)是分子振動(dòng)光譜合頻、一級(jí)倍頻和二級(jí)倍頻的吸收譜帶,吸收峰較強(qiáng)。另外,前期研究[27]表明,果蔬表面的顏色和果肉的顏色在短波近紅外區(qū)對(duì)檢測(cè)精度具有顯著性影響,而對(duì)長(zhǎng)波近紅外光譜影響較小。綜合性能和價(jià)格因素,本研究選擇美國(guó)Texas Instruments公司的DLP2010NIR DMD模塊用于系統(tǒng)設(shè)計(jì)。近紅外光譜儀內(nèi)部光路與信號(hào)傳輸過(guò)程如圖2所示。通過(guò)果蔬樣本的漫反射光,經(jīng)入射狹縫進(jìn)入光譜儀,首先通過(guò)準(zhǔn)直透鏡和885 nm的長(zhǎng)波通濾光片,然后經(jīng)衍射光柵反射將光分成若干波長(zhǎng)下的光,再經(jīng)聚焦透鏡投射到DMD上;通過(guò)嵌入式處理器精確控制DMD中的每一個(gè)微鏡,在每一瞬間僅有特定波長(zhǎng)的光傳輸?shù)教綔y(cè)器。選用InGaAs單點(diǎn)探測(cè)器獲取光電信號(hào),同時(shí)配置半導(dǎo)體制冷器以提高信號(hào)的精度和穩(wěn)定性。另外,在狹縫前設(shè)有聚焦透鏡,可有效獲取2.5 mm視窗內(nèi)的漫反射光;通過(guò)調(diào)整狹縫的大小可以調(diào)節(jié)光譜分辨率和信噪比。通過(guò)試驗(yàn)發(fā)現(xiàn),所設(shè)計(jì)的系統(tǒng)可有效獲取900~ 1 700 nm范圍的近紅外光譜。

1. 果蔬樣本 2. 光源 3. 狹縫 4. 準(zhǔn)直透鏡 5. 衍射光柵 6. 聚焦透鏡 7. 數(shù)字微鏡器件 8. 采集透鏡 9. 單點(diǎn)探測(cè)器 10. 半導(dǎo)體制冷器 11. 信號(hào)放大器 12. 模數(shù)轉(zhuǎn)換器 13. 嵌入式處理器
1.2 微型光源
近紅外光譜檢測(cè)系統(tǒng)需要提供寬波段的光源,一般選用鹵鎢燈。手持式檢測(cè)系統(tǒng)設(shè)計(jì)中光源在滿足基本能量和強(qiáng)度要求的前提下要求尺寸小、功耗低。在低功耗微型光源方面,設(shè)置前置透鏡可以將光有效聚集到特定方向,在光源前部可以獲得10倍于同等功耗的普通光源的光強(qiáng)度。光源的這種設(shè)計(jì)發(fā)散性得到收斂,具有一定的焦距和工作平面,在工作面上光分布需要優(yōu)化調(diào)試,包括光強(qiáng)度、光照均勻性和一致性等。光源首選C-6結(jié)構(gòu)的細(xì)鎢絲,具有低電壓高電流的特點(diǎn),采用高品質(zhì)玻璃封裝,色溫2 200 K,使用壽命25 000 h。本研究選用美國(guó)ILT公司的L1005型前置透鏡微型精密光源。近紅外光譜的采集設(shè)計(jì)為反射模式,2個(gè)光源對(duì)稱布局,光源的主軸與入射狹縫平面呈40°夾角,光投射到果蔬表面進(jìn)入內(nèi)部,然后光經(jīng)體反射傳輸?shù)焦庾V儀,可以避免無(wú)效鏡面反射光進(jìn)入光譜儀。
1.3 控制與通訊模式
手持式近紅外光譜檢測(cè)系統(tǒng)中DMD的控制的是獲取高質(zhì)量光譜的關(guān)鍵,由DLPC 150微控制器可編程實(shí)現(xiàn)DLP2010NIR DMD的精確控制。單點(diǎn)InGaAs探測(cè)器的信號(hào)經(jīng)低信噪比差分放大器,傳到串行外設(shè)接口的模數(shù)轉(zhuǎn)換器,最后傳到嵌入式處理器,如圖2所示。嵌入式處理器設(shè)有低能耗藍(lán)牙模塊和微型USB連接器,在后續(xù)軟件開(kāi)發(fā)過(guò)程,開(kāi)發(fā)雙通訊模式,滿足不同用戶的多功能需求。
1.4 檢測(cè)軟件設(shè)計(jì)
根據(jù)檢測(cè)的需要,自行開(kāi)發(fā)果蔬品質(zhì)手持式近紅外光譜檢測(cè)軟件。軟件設(shè)計(jì)過(guò)程采取模塊化結(jié)構(gòu)設(shè)計(jì),即根據(jù)功能要求將整個(gè)系統(tǒng)劃分為不同的功能模塊,便于維護(hù)和功能擴(kuò)展。果蔬品質(zhì)檢測(cè)模型內(nèi)置于檢測(cè)軟件中,采用多線程編程技術(shù)實(shí)現(xiàn),可以一機(jī)多用,用于檢測(cè)多種果蔬。根據(jù)不同用戶的需要,基于Windows系統(tǒng)和Android平臺(tái)開(kāi)發(fā)專用軟件,實(shí)現(xiàn)USB和藍(lán)牙雙通訊功能。程序初始化完成,打開(kāi)開(kāi)始采集即進(jìn)入工作狀態(tài),操作簡(jiǎn)單。
鑒于番茄具有水果和蔬菜的雙重屬性,具有一定的代表性,已有研究表明,近紅外光譜檢測(cè)番茄的品質(zhì)是可行[28-29]。本研究選擇番茄為研究對(duì)象驗(yàn)證設(shè)計(jì)的果蔬品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)的性能。
2.1 番茄品質(zhì)測(cè)定
番茄中的可溶性固形物含量(soluble solids content, SSC)是反映番茄主要營(yíng)養(yǎng)物質(zhì)含量重要指標(biāo)之一,也以其作為評(píng)價(jià)番茄及其制品品質(zhì)的重要指標(biāo)之一。番茄SSC測(cè)定參考GB12295[30]采用折射儀法,測(cè)量?jī)x器為WYA-2S數(shù)字阿貝折射儀。取番茄上對(duì)應(yīng)光譜采集的部位,經(jīng)多層紗布過(guò)濾擠汁滴在折光儀鏡面,測(cè)定溫度修正為20 ℃的白利度(%)。
番茄紅素是成熟番茄的主要色素,也是評(píng)價(jià)番茄品質(zhì)等級(jí)的關(guān)鍵指標(biāo)。番茄紅素含量(lycopene content, LC)的測(cè)定參考GB 28316-2012[31],用二氯甲烷浸提番茄紅素,溶液中加入2,6-二叔丁基對(duì)甲酚用來(lái)防止番茄紅素氧化。用紫外-可見(jiàn)光分光光度計(jì)在472 nm的最大吸收波長(zhǎng)處測(cè)定吸光度,計(jì)算番茄紅素的質(zhì)量分?jǐn)?shù),單位為mg/kg。
近紅外光譜具有多組分同時(shí)檢測(cè)的特點(diǎn),本研究采用手持式近紅外光譜檢測(cè)系統(tǒng)同時(shí)檢測(cè)番茄中的可溶性固形物含量和番茄紅素。測(cè)試番茄樣本共78個(gè),按照約2∶1的比例劃分為校正集與預(yù)測(cè)集,品質(zhì)指標(biāo)的測(cè)定結(jié)果統(tǒng)計(jì)如表1所示,發(fā)現(xiàn)預(yù)測(cè)集樣本的統(tǒng)計(jì)結(jié)果與校正集樣本的范圍、均值和標(biāo)準(zhǔn)偏差相當(dāng),樣本選擇具有一定的達(dá)標(biāo)性。

表1 番茄品質(zhì)測(cè)定結(jié)果
2.2 番茄近紅外光譜采集與分析
番茄品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)經(jīng)優(yōu)化采集參數(shù),設(shè)置波長(zhǎng)范圍為900~1 700 nm,光譜分辨率為4.68 nm,掃描點(diǎn)數(shù)400,掃描次數(shù)為3。78個(gè)番茄樣本光譜采集完成獲得78×400的光譜矩陣,原始近紅外光譜如圖3所示。
從圖3可以看出,每條光譜之間都是相似的變化趨勢(shì),無(wú)法直接進(jìn)行分組或分類。因番茄具有很高的含水率,在1 450 nm有很強(qiáng)的水分吸收峰(水分一倍頻的伸縮振動(dòng)),在970 nm是水分的二倍頻吸收峰。在1 120~1 260 nm對(duì)應(yīng)CH鍵的二倍頻吸收譜帶,在1 350~1 480 nm是CH鍵、OH鍵和NH鍵一倍頻的復(fù)合吸收[32]。其中番茄中的番茄紅素是一種不飽和烯烴化合物,由11個(gè)不飽和共軛雙鍵和兩個(gè)非共軛不飽和雙鍵組成的長(zhǎng)鏈脂肪烴[33]。獲取的900~1 700 nm范圍的近紅外光譜包括CH鍵一倍頻和二倍頻吸收的信息,可以用來(lái)預(yù)測(cè)番茄紅素的含量,同樣也可預(yù)測(cè)番茄中的可溶性固形物含量。

注:A=log(1/R)為近紅外光譜的吸光度,其中R為近紅外光譜的反射率。
近紅外光譜采集過(guò)程會(huì)夾入大量基線漂移、高頻隨機(jī)噪聲、光散射等噪聲信息,導(dǎo)致近紅外光譜與樣品內(nèi)有效成分含量間關(guān)系受到干擾,直接影響模型的魯棒性。在比較多種光譜預(yù)處理方法的基礎(chǔ)上,發(fā)現(xiàn)標(biāo)準(zhǔn)正態(tài)變量變換(standard normal variate transformation, SNV)預(yù)處理方法可以提高模型的穩(wěn)定性。究其原因是SNV方法可以消除物料表面光散射及光程變化對(duì)漫反射光信號(hào)的影響。
2.3 番茄品質(zhì)模型的建立
考慮到近紅外光譜帶重疊、吸收信號(hào)弱的特點(diǎn),在番茄品質(zhì)模型的建立過(guò)程,首先選擇特征波段,去除無(wú)信息變量和相關(guān)度不高的波段,然后再利用特征波長(zhǎng)選擇方法優(yōu)選少量的特征波長(zhǎng),消除光譜數(shù)據(jù)內(nèi)部存在的共線性關(guān)系,降低模型計(jì)算量,同時(shí)簡(jiǎn)化模型,提高模型的質(zhì)量。本研究先利用聯(lián)合區(qū)間偏最小二乘法(synergy internal partial least square, siPLS)選擇特征波段,再利用連續(xù)投影算法(successive projections algorithm, SPA)選擇特征波長(zhǎng)[34-35],分別建立番茄可溶性固形物含量和番茄紅素的定量分析模型。在模型建立過(guò)程中,以校正相關(guān)系數(shù)(correlation coefficient of calibration,R)、預(yù)測(cè)相關(guān)系數(shù)(correlation coefficient of prediction,R)、校正均方根誤差(root mean square error of calibration, RMSEC)和預(yù)測(cè)均方根誤差(root mean squared error of prediction, RMSEP)評(píng)價(jià)模型性能。
應(yīng)用siPLS建模時(shí),將整個(gè)光譜區(qū)域劃分為20個(gè)子區(qū)間,聯(lián)合4個(gè)子區(qū)間優(yōu)選特征波段,然后在選擇的特征波段上利用連續(xù)投影算法選擇特征波長(zhǎng)。對(duì)可溶性固形物含量,優(yōu)選9個(gè)特征波長(zhǎng),分別是1 148.85、1 298.43、1 141.73、1 319.80、1 123.92、1 373.22、1 134.61、1 355.42和1 391.03 nm。可溶性固形物含量選擇的特征譜區(qū)和優(yōu)選的特征波長(zhǎng)如圖4a所示。對(duì)番茄紅素,優(yōu)選11個(gè)特征波長(zhǎng),分別是1 141.73、1 131.05、1 138.17、1 198.71、 1 159.54、1 145.29、1 120.36、1 148.85、1 319.80、1 113.24和1 348.29 nm,番茄紅素模型選擇的特征譜區(qū)和優(yōu)選的特征波長(zhǎng)如圖4b所示。

圖4 可溶性固形物含量和番茄紅素的優(yōu)選的光譜區(qū)間和選擇的特征波長(zhǎng)
番茄可溶性固形物含量和番茄紅素模型建立結(jié)果見(jiàn)表2,所建立的聯(lián)合區(qū)間偏最小二乘模型選用的變量,對(duì)SSC和LC,從全譜的400個(gè)變量分別減少為79個(gè)和81個(gè)。采用連續(xù)投影算法優(yōu)選更少的變量而獲得了與聯(lián)合區(qū)間偏最小二乘法相當(dāng)?shù)念A(yù)測(cè)精度。可見(jiàn)前向循環(huán)變量選擇的連續(xù)投影算法可以從共線性的光譜變量中有效提取信息。可見(jiàn)采用先選擇特征光譜波段再優(yōu)選特征波長(zhǎng)的策略可有效檢測(cè)番茄的品質(zhì)指標(biāo)。

表2 番茄品質(zhì)指標(biāo)不同校正模型建立結(jié)果
1)根據(jù)果蔬快速無(wú)損檢測(cè)低成本、實(shí)用化和小型化的實(shí)際產(chǎn)業(yè)需求,構(gòu)建了手持式近紅外光譜檢測(cè)系統(tǒng)。采用數(shù)字微鏡器件結(jié)合點(diǎn)陣探測(cè)器的組合方式在保證精度的同時(shí)顯著降低了儀器成本,同時(shí)優(yōu)化設(shè)計(jì)了微型光源、控制與通訊模塊,開(kāi)發(fā)了雙通訊模式的專用檢測(cè)軟件,有效獲取900~1 700 nm的近紅外光譜。
2)選擇番茄為果蔬的代表,果蔬品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)經(jīng)優(yōu)化采集參數(shù),對(duì)采集的光譜進(jìn)行解析,提取1 120~1 480 nm范圍的光譜可以表征番茄的可溶性固形物含量和番茄紅素等品質(zhì)指標(biāo)。
3)研究先采用聯(lián)合區(qū)間偏最小二乘法選擇特征波段,再利用連續(xù)投影算法選擇特征波長(zhǎng)的模型簡(jiǎn)化策略,分別建立番茄可溶性固形物含量和番茄紅素的定量分析模型。可溶性固形物含量模型的預(yù)測(cè)相關(guān)系數(shù)R和預(yù)測(cè)均方根誤差RMSEP分別為0.899和0.133%;番茄紅素模型的預(yù)測(cè)相關(guān)系數(shù)Rp和預(yù)測(cè)均方根誤差RMSEP分別為0.886和2.508 mg/kg。
研究結(jié)果表明,研發(fā)的手持式近紅外光譜檢測(cè)系統(tǒng)可以實(shí)現(xiàn)果蔬主要品質(zhì)指標(biāo)的快速無(wú)損檢測(cè)。研究為低成本、實(shí)用化、小型化近紅外光譜檢測(cè)系統(tǒng)設(shè)計(jì)提供方法參考。
[1] Zhou J, Kai L I, Liang Q. Food safety controls in different governance structures in China’s vegetable and fruit industry[J]. Journal of Integrative Agriculture, 2015, 14(11): 2189-2202.
[2] 王偉新,向云,祁春節(jié). 中國(guó)水果產(chǎn)業(yè)地理集聚研究:時(shí)空特征與影響因素[J]. 經(jīng)濟(jì)地理,2013,33(8):97-103. Wang Weixin, Xiang Yun, Qi Chunjie. Study on geographic agglomeration of fruit industry in china: Spatial-temporal characteristics and affecting factors[J]. Economic Geography, 2013, 33(8): 97-103. (in Chinese with English abstract)
[3] Ernest Liu K, Chang H H, Chern W S. Examining changes in fresh fruit and vegetable consumption over time and across regions in urban China[J]. China Agricultural Economic Review, 2011, 3(3): 276-296.
[4] Zhao Jiewen, Chen Quansheng, Saritporn Vittayapadung, et al. Determination of apple firmness using hyperspectral imaging technique and multivariate calibrations[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2009, 25(11): 226-231.
[5] Blakey R J. Evaluation of avocado fruit maturity with a portable near-infrared spectrometer[J]. Postharvest Biology and Technology, 2016, 121: 101-105.
[6] Guo W, Fang L, Liu D, et al. Determination of soluble solids content and firmness of pears during ripening by using dielectric spectroscopy[J]. Computers and Electronics in Agriculture, 2015, 117: 226-233.
[7] 傅霞萍,應(yīng)義斌. 基于NIR和Raman光譜的果蔬質(zhì)量檢測(cè)研究進(jìn)展與展望[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2013,44(8):148-164. Fu Xiaping, Ying Yibin. Application of NIR and Raman spectroscopy for quality and safety inspection of fruits and vegetables: a review[J]. Transactions of the Chinese Society for Agricultural Machinery, 2013, 44(8): 148-164. (in Chinese with English abstract)
[8] Xie L, Wang A, Xu H, et al. Applications of near-infrared systems for quality evaluation of fruits: a review[J]. Transactions of the ASABE, 2016, 59(2): 399-419.
[9] 褚小立,陸婉珍. 近五年我國(guó)近紅外光譜分析技術(shù)研究與應(yīng)用進(jìn)展[J]. 光譜學(xué)與光譜分析,2014,34(10):2595-2605. Chu Xiaoli, Lu Wanzhen. Research and application progress of near infrared spectroscopy analytical technology in China in the past five years[J]. Spectroscopy and Spectral Analysis, 2014, 34(10): 2595-2605. (in Chinese with English abstract)
[10] Fernández-Espinosa A J. Combining PLS regression with portable NIR spectroscopy to on-line monitor quality parameters in intact olives for determining optimal harvesting time[J]. Talanta, 2016, 148: 216-228.
[11] 劉燕德,吳明明,孫旭東,等. 黃桃表面缺陷和可溶性固形物光譜同時(shí)在線檢測(cè)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2016,32(6):289-295. Liu Yande, Wu Mingming, Sun Xudong, et al. Simultaneous detection of surface deficiency and soluble solids content for Amygdalus persica by online visible near infrared transmittance spectroscopy[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016, 32(6): 289-295. (in Chinese with English abstract)
[12] 郭志明,黃文倩,陳全勝,等. 蘋果腐心病的透射光譜在線檢測(cè)系統(tǒng)設(shè)計(jì)及試驗(yàn)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2016,32(6):283-288. Guo Zhiming, Huang Wenqian, Chen Quansheng, et al. Design and test of on line detection system for apple core rot disease based on transmitted spectrum[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016, 32(6): 283-288. (in Chinese with English abstract)
[13] Ruslan R, Ehsani R, Lee W S. Quantification of total soluble solids and titratable acidity for citrus maturity using portable vis-NIR spectroradiometer[J]. Applied Engineering in Agriculture, 2012, 28(5): 735-743.
[14] Paula J T, Faria J T V, Marcos V, et al. Physicochemical characteristics and bioactive compounds in tomato fruits harvested at different ripening stages[J]. Horticultura Brasileira, 2015, 33(4): 434-440.
[15] Yuan L, Cai J, Sun L, et al. Nondestructive measurement of soluble solids content in apples by a portable fruit analyzer[J]. Food Analytical Methods, 2016, 9(3): 785-794.
[16] Cirilli M, Bellincontro A, Urbani S, et al. On-field monitoring of fruit ripening evolution and quality parameters in olive mutants using a portable NIR-AOTF device[J]. Food chemistry, 2016, 199: 96-104.
[17] Pérez-Marín D, Paz P, Guerrero J E, et al. Miniature handheld NIR sensor for the on-site non-destructive assessment of post-harvest quality and refrigerated storage behavior in plums[J]. Journal of food engineering, 2010, 99(3): 294-302.
[18] 樊書祥,黃文倩,李江波,等. 特征變量?jī)?yōu)選在蘋果可溶性固形物近紅外便攜式檢測(cè)中的應(yīng)用[J]. 光譜學(xué)與光譜分析,2014,34(10):2707-2712. Fan Shuxiang, Huang Wenqian, Li Jiangbo, et al. Application of characteristic NIR variables selection in portable detection of soluble solids content of apple by near infrared spectroscopy[J]. Spectroscopy and Spectral Analysis, 2014, 34(10): 2707-2712. (in Chinese with English abstract)
[19] Kafle G K, Khot L R, Jarolmasjed S, et al. Robustness of near infrared spectroscopy based spectral features for non- destructive bitter pit detection in honeycrisp apples[J]. Postharvest Biology and Technology, 2016, 120: 188-192.
[20] Yu X, Lu Q, Gao H, et al. Development of a handheld spectrometer based on a linear variable filter and a complementary metal-oxide-semiconductor detector for measuring the internal quality of fruit[J]. Journal of Near Infrared Spectroscopy, 2016, 24(1): 69-76.
[21] Civelli R, Giovenzana V, Beghi R, et al. A simplified, light emitting diode (LED) based, modular system to be used for the rapid evaluation of fruit and vegetable quality: Development and validation on dye solutions[J]. Sensors, 2015, 15(9): 22705-22723.
[22] Beghi R, Giovenzana V, Civelli R, et al. Setting-up of a simplified handheld optical device for decay detection in fresh-cut Valerianella locusta L[J]. Journal of Food Engineering, 2014, 127: 10-15.
[23] Dos Santos C A T, Lopo M, Páscoa R N M J, et al. A review on the applications of portable near-infrared spectrometers in the agro-food industry[J]. Applied spectroscopy, 2013, 67(11): 1215-1233.
[24] Sánchez M T, De la Haba M J, Pérez-Marín D. Internal and external quality assessment of mandarins on-tree and at harvest using a portable NIR spectrophotometer[J]. Computers and electronics in agriculture, 2013, 92: 66-74.
[25] Spencer A P, Spokoyny B, Ray S, et al. Mapping multidimensional electronic structure and ultrafast dynamics with single-element detection and compressive sensing[J]. Nature Communications, 2016, 7: 10434.
[26] Rueda H, Arguello H, Arce G R. DMD-based implementation of patterned optical filter arrays for compressive spectral imaging[J]. Journal of the Optical Society of America A-Optics Image Science and Vision, 2015, 32(1): 80-89.
[27] Guo Z M, Huang W Q, Peng Y K, et al Color compensation and comparison of shortwave near infrared and long wave near infrared spectroscopy for determination of soluble solids content of ‘Fuji’ apple[J]. Postharvest Biology and Technology, 2016, 115: 81-90.
[28] Ding X, Guo Y, Ni Y, et al. A novel NIR spectroscopic method for rapid analyses of lycopene, total acid, sugar, phenols and antioxidant activity in dehydrated tomato samples[J]. Vibrational Spectroscopy, 2016, 82: 1-9.
[29] Ayvaz H, Sierra-Cadavid A, Aykas D P, et al. Monitoring multicomponent quality traits in tomato juice using portable mid-infrared (MIR) spectroscopy and multivariate analysis[J]. Food Control, 2016, 66: 79-86.
[30] 水果、蔬菜制品可溶性固形物含量的測(cè)定--折射儀法,GB12295-1990[S].
[31] 食品安全國(guó)家標(biāo)準(zhǔn)食品添加劑番茄紅,GB 28316-2012 [S]. 北京:中國(guó)標(biāo)準(zhǔn)出版社,2012.
[32] Zhu Q, He C, Lu R, et al. Ripeness evaluation of ‘Sun Bright’tomato using optical absorption and scattering properties[J]. Postharvest Biology and Technology, 2015, 103: 27-34.
[33] De Nardo T, Shiroma-Kian C, Halim Y, et al. Rapid and simultaneous determination of lycopene and β-carotene contents in tomato juice by infrared spectroscopy[J]. Journal of agricultural and food chemistry, 2009, 57(4): 1105-1112.
[34] Wu D, He Y, Nie P, et al. Hybrid variable selection in visible and near-infrared spectral analysis for non-invasive quality determination of grape juice[J]. Analytica Chimica Acta, 2010, 659(1): 229-237.
[35] Soares S F C, Gomes A A, Araujo M C U, et al. The successive projections algorithm[J]. TrAC Trends in Analytical Chemistry, 2013, 42: 84-98.
Design and experiment of handheld near-infrared spectrometer for determination of fruit and vegetable quality
Guo Zhiming1,2, Chen Quansheng1※, Zhang Bin1, Wang Qingyan2, Ouyang Qin1, Zhao Jiewen1
(1.212013,;2.100097,)
Determination of internal quality of fruit and vegetable with a suitable technique is crucial for processing detection and quality control. While substantial progress has recently been made in the miniaturization of near-infrared (NIR) spectrometers, there remains continued interest from end-users and product developers in pushing the technology envelope toward even smaller and lower cost analyzers. The potential of these instruments to revolutionize on-site applications can be realized only if the reduction in size does not compromise performance of the spectrometer beyond the practical need of a given application. In this paper, the working principle of a novel, extremely miniaturized NIR spectrometer is presented. The ultra-compact spectrometer relies on digital micromirror device (DMD) technology for the light dispersing element. DMD is a two-dimensional array of electro-mechanical mirror elements whose surface normal angles can be controlled. Digitally programmable DMD can set the spectral resolution and wavelength range according to user needs, adjust the integration time, and adapt the luminous flux. The system design with DMD and single-pixel InGaAs detector can significantly reduce the cost, and meanwhile ensure the detect precision. The DLP (digital light procession) NIRscan module is used as spectrometer optical engine in the miniaturized system. In the specific implementation, a sample is placed against the sapphire front window of the reflectance head. During a scan, the sample absorbs a specific amount of NIR light and diffusely reflects the non-absorbed light into the system. The illuminating lamps are designated as lens-end lamps because the front end of the glass bulb is formed into a lens that directs more lights from the filament to the sample test region. The collection lens gathers collimated light from a 2.5 mm diameter region at the sample window. The handheld system supports the following modes of operation: USB (Universal Serial Bus) connection and Bluetooth for 2 communication channels. Special analyzer software was developed for quality inspection based on multithread programming technology. The advantages of this software are presented by the process of modular design, including software system initialization, information communication, information interaction, spectral data acquisition and processing, spectral curve real-time display, quality index calculation, and statistics and save of detection results. Miniaturized handheld NIR spectrometer was developed and used to acquire reflectance spectra from fruit and vegetable samples in the wavelength range of 900-1 700 nm. In order to verify the design and performance, tomato was selected as research object. A total of 78 tomato samples were randomly divided into 2 subsets. The first subset was called the calibration set with 52 samples and used for building model, while the other one was called the prediction set with 26 samples and used for testing the robustness of the model. In the process of model establishment, a simplified strategy was proposed. Firstly, characteristic spectrum bands were selected to remove the uninformative variable and the low-correlation band. And then feature wavelengths were optimized to eliminate the collinearity relationship in the spectral data. Finally, simplified model was built, which had good robustness and stability. Synergy internal partial least square (siPLS) and successive projections algorithm (SPA) were sequentially applied to calibrate models. The siPLS was applied to select an optimized spectral interval and an optimized combination of spectral regions selected from informative regions in model calibration. The subsequent application of SPA to this reduced domain could lead to an efficient and refined model. The measurement results of the final model were achieved as follows: correlation coefficient (R) was 0.899 and root mean square error of prediction (RMSEP) was 0.133% for soluble solid content in tomato, andRwas 0.886 and RMSEP was 2.508 mg/kg for lycopene content in tomato. The results will provide the method reference for rapid, non-destructive, and on-site detection technology and equipment of fruit internal quality.
nondestructive detection; spectrum analysis; agriculture products; near infrared spectroscopy; handheld detection system; digital micromirror device; fruit and vegetable quality
10.11975/j.issn.1002-6819.2017.08.033
TP274.4;O433.4
A
1002-6819(2017)-08-0245-06
2016-08-23
2016-11-06
國(guó)家自然科學(xué)基金(31501216);國(guó)家科技支撐計(jì)劃(2015BAD19B03);中國(guó)博士后科學(xué)基金(2016M600379);江蘇省高校自然科學(xué)研究面上項(xiàng)目(16KJB550002);江蘇省博士后科研資助計(jì)劃(1601080B);食品安全大數(shù)據(jù)技術(shù)北京市重點(diǎn)實(shí)驗(yàn)室開(kāi)放課題(BKBD- 2016KF06);江蘇大學(xué)高級(jí)人才基金(15JDG169)
郭志明,博士,講師,主要從事食品品質(zhì)安全的光電檢測(cè)技術(shù)與裝備研究。鎮(zhèn)江 江蘇大學(xué)食品與生物工程學(xué)院,212013。 Email:zhmguo@126.com
陳全勝,教授,博士生導(dǎo)師,主要從事農(nóng)產(chǎn)品無(wú)損檢測(cè)技術(shù)研究。鎮(zhèn)江 江蘇大學(xué)食品與生物工程學(xué)院,212013。Email:qschen@ujs.edu.cn
郭志明,陳全勝,張 彬,王慶艷,歐陽(yáng)琴,趙杰文. 果蔬品質(zhì)手持式近紅外光譜檢測(cè)系統(tǒng)設(shè)計(jì)與試驗(yàn)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2017,33(8):245-250. doi:10.11975/j.issn.1002-6819.2017.08.033 http://www.tcsae.org
Guo Zhiming, Chen Quansheng, Zhang Bin, Wang Qingyan, Ouyang Qin, Zhao Jiewen. Design and experiment of handheld near-infrared spectrometer for determination of fruit and vegetable quality[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(8): 245-250. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2017.08.033 http://www.tcsae.org