How Can Spectral Technology Non-Destructively Determine Fruit Ripeness?

想(xiǎng)知道牛油果(guǒ)何(hé)時入口最(zuì)佳(jiā)，榴(liú)蓮是否熟透(tòu)？水果成熟度(dù)檢(jiǎn)測一(yī)直(zhí)是農業領(lǐng)域的重要課(kè)題，本篇(piān)將(jiāng)介紹光(guāng)譜和高光(guāng)譜(pǔ)成像(xiàng)技(jì)術(shù)如何(hé)為(wéi)這一課(kè)題(tí)提(tí)供了(liǎo)創新解決方(fāng)案，通過無(wú)損檢測實現精(jīng)準判斷。

水果(guǒ)成熟過程中，其內(nèi)部化學成分（如葉綠素
，類胡蘿卜素
，糖分
，酸(suān)度(dù)等(děng)）會(huì)發(fā)生(shēng)規律性(xìng)變(biàn)化(huà)，這(zhè)些(xiē)物質對特定波(bō)長的光(guāng)具(jù)有(yǒu)吸收(shōu)和反射(shè)特性(xìng)。研究表明(míng)，可(kě)見(jiàn)光(guāng)波段主(zhǔ)要(yào)反映色素變化(huà)，而近紅外區(qū)域則與水分
，糖分等(děng)內(nèi)部成(chéng)分密切相關。

在實際應(yīng)用中(zhōng)，不同水果種(zhǒng)類因其(qí)生(shēng)理特(tè)性差異，需要采用(yòng)特定的特征(zhēng)波長和(hé)不同(tóng)的算法模(mó)型。

Want to know when an avocado is at its peak or if a durian is perfectly ripe? Fruit ripeness detection has always been a critical topic in agriculture. This article explores how spectral and hyperspectral imaging technologies provide innovative solutions for this challenge, enabling precise judgment through non-destructive testing.

During fruit ripening, internal chemical components (such as chlorophyll, carotenoids, sugars, acidity, etc.) undergo regular changes. These substances exhibit unique absorption and reflection characteristics for specific wavelengths of light. Research shows that the visible light spectrum primarily reflects pigment changes, while the near-infrared region is closely related to internal components like moisture and sugar content.

In practical applications, different fruit types require specific characteristic wavelengths and distinct algorithm models due to variations in their physiological properties.

小(xiǎo)果的(dí)成熟(shú)度(dù)分析 / How Can Spectral Technology Non-Destructively Determine Fruit Ripeness?

一(yī)項甜(tián)橙研究采用400-1000nm波(bō)段的(dí)可(kě)見/近(jìn)紅外光(guāng)譜，結(jié)合偏最小(xiǎo)二乘法(fǎ)(PLS)，成(chéng)功(gōng)預(yù)測了可溶(róng)性固形物
，可(kě)滴定(dìng)酸和維生素C含量，為成(chéng)熟(shú)期預(yù)測提供(gōng)了量化依(yī)據(jù)。

香(xiāng)蕉(jiāo)成熟(shú)度檢測常利(lì)用高(gāo)光譜(pǔ)成(chéng)像技(jì)術在400-1000nm範圍內(nèi)采(cǎi)集數據(jù)。一個研(yán)究團隊通過主成分(fēn)分(fēn)析(xī)(PCA)結合極(jí)限學習(xí)機(ELM)建立的模(mó)型(xíng)，對可溶性固(gù)形物和硬(yìng)度的預測相關(guān)係數R²分(fēn)別達到(dào)0.92和0.94。

針(zhēn)對(duì)牛(niú)油(yóu)果的(dí)研究(jiū)發(fā)現，其成(chéng)熟(shú)度(dù)判斷主(zhǔ)要依賴於800nm以上(shàng)的(dí)近紅(hóng)外(wài)信息(xī)，而520~650nm的(dí)可(kě)見(jiàn)光範圍(wéi)則(zé)有助於區(qū)分未(wèi)成熟(shú)與成(chéng)熟果實(shí)。研究人員(yuán)開(kāi)發(fā)的高光(guāng)譜(pǔ)卷積(jī)神經(jīng)網絡(luò)(HS-CNN)模型，在牛油果成(chéng)熟度(dù)分類中準確(què)率(shuài)超(chāo)過(guò)90%。

A study on sweet oranges utilized visible/near-infrared spectroscopy in the 400–1000 nm range, combined with partial least squares (PLS), to successfully predict soluble solids, titratable acidity, and vitamin C content, providing a quantitative basis for ripening stage prediction.

For banana ripeness detection, hyperspectral imaging technology is often employed to collect data within the 400–1000 nm range. One research team developed a model using principal component analysis (PCA) combined with an extreme learning machine (ELM), achieving prediction correlation coefficients (R²) of 0.92 and 0.94 for soluble solids and firmness, respectively.

Research on avocados found that ripeness determination primarily relies on near-infrared information above 800 nm, while the visible light range of 520–650 nm helps distinguish unripe from ripe fruit. A hyperspectral convolutional neural network (HS-CNN) model developed by researchers achieved over 90% accuracy in avocado ripeness classification.

高(gāo)光(guāng)譜數據對牛油(yóu)果(guǒ)的(dí)成熟度(dù)分類的決(jué)策影響：牛油果的空間維(wéi)
，光(guāng)譜維(wéi)圖(tú)像(xiàng) / The impact of the input on the decision of the class for an avocado

皮(pí)厚且(qiě)堅(jiān)硬(yìng)的(dí)水(shuǐ)果，如何(hé)檢(jiǎn)測？ / How to Detect Ripeness in Thick-Skinned, Hard Fruits?

對於(yú)西瓜(guā)
，哈密瓜(guā)
，榴(liú)蓮(lián)等(děng)皮厚且(qiě)堅硬的水果，成熟(shú)度檢測麵臨(lín)很大的挑(tiāo)戰(zhàn)。

一項西瓜研(yán)究使用了(liǎo)近紅外光(guāng)譜(pǔ)（NIRS）技術(shù)，涉及908~1676nm和(hé)950~1650nm光譜(pǔ)範(fàn)圍，檢(jiǎn)測(cè)了249個完整西瓜（152個淺(qiǎn)綠條(tiáo)紋果(guǒ)皮(pí)，97個深綠(lǜ)純色(sè)果皮(pí)）。利用偏最小二乘(chéng)判別(bié)分(fēn)析（PLS-DA），構(gòu)建(jiàn)可溶性固(gù)形(xíng)物含量（SSC）的定(dìng)量(liáng)模(mó)型(xíng)。結(jié)果顯(xiǎn)示(shì)，淺綠條(tiáo)紋和深綠(lǜ)純色西(xī)瓜(guā)的正確分類率分別為(wéi)66.4%和82.2%，針(zhēn)對不(bù)同類(lèi)型西(xī)瓜(guā)分(fēn)別建立模型(xíng)能獲(huò)得更好(hǎo)結果。

哈(hā)密瓜與(yǔ)西瓜(guā)類似，研(yán)究顯示(shì)，其(qí)可溶(róng)性(xìng)固(gù)形物(wù)含(hán)量(liáng)與(yǔ)特(tè)定(dìng)波(bō)長(cháng)反(fǎn)射率(shuài)存在(zài)強相關(guān)性(xìng)。通(tōng)過優(yōu)化選擇(zé)的特征(zhēng)波長建立的(dí)簡(jiǎn)化模(mó)型(xíng)，既(jì)保(bǎo)持了預測精(jīng)度(dù)，又提高了(liǎo)檢測速度(dù)。

榴蓮(lián)作為巨(jù)大挑戰性(xìng)的厚皮(pí)水果(guǒ)之一，其(qí)成熟(shú)度檢(jiǎn)測(cè)一直依(yī)賴經驗判斷(duàn)或(huò)破壞性方法。榴蓮(lián)成熟度檢測(cè)常(cháng)依賴(lài)經驗判斷或破壞(huài)性方(fāng)法。一(yī)項研究采用1100~2500nm光譜範圍，使(shǐ)用果(guǒ)皮和(hé)莖(jīng)的(dí)光(guāng)譜信息對(duì)果(guǒ)肉幹(gān)物質，進行間接預測成熟(shú)度。

研(yán)究發(fā)現(xiàn)，在將榴蓮(lián)分為未成熟(shú)
，早成熟和成熟類(lèi)別的過(guò)程(chéng)中(zhōng)，外(wài)皮模型(xíng)更優；預(yù)測(cè)幹物(wù)質(zhì)含(hán)量(liáng)方(fāng)麵，果皮(pí)模型(xíng)表現更好(hǎo)。研究(jiū)人(rén)員發現(xiàn)，盡(jìn)管(guǎn)與參(cān)考(kǎo)果肉模型(xíng)的精度(dù)相(xiāng)比，準確(què)度(dù)相對(duì)較低(dī)，但在選定波長(cháng)下，組合(hé)分(fēn)析(xī)外皮(pí)和(hé)莖幹(gān)光譜(pǔ)數(shù)據可提(tí)供較(jiào)高(gāo)分類精度。

A watermelon study employed near-infrared spectroscopy (NIRS) technology, covering spectral ranges of 908–1676 nm and 950–1650 nm, to examine 249 intact watermelons (152 with light green striped rinds and 97 with dark green solid rinds). Using partial least squares discriminant analysis (PLS-DA), a quantitative model for soluble solids content (SSC) was constructed. Results showed correct classification rates of 66.4% for light green striped watermelons and 82.2% for dark green solid ones, indicating that separate models for different types yield better outcomes.

Similar to watermelons, cantaloupe studies revealed strong correlations between soluble solids content and reflectance at specific wavelengths. Simplified models built with optimized characteristic wavelengths maintained prediction accuracy while improving detection speed.

Durian, one of the most challenging thick-skinned fruits, has traditionally relied on experiential judgment or destructive methods for ripeness assessment. A study used the 1100–2500 nm spectral range, leveraging rind and stem spectral data to indirectly predict pulp dry matter content as an indicator of ripeness.

The study found that for classifying durians into unripe, early ripe, and ripe categories, the rind model performed better. In predicting dry matter content, the rind model also showed superior performance. Researchers noted that although the accuracy was relatively lower compared to reference pulp models, combining rind and stem spectral data at selected wavelengths could achieve higher classification precision.

西(xī)瓜研究 / Watermelon Study：

西(xī)瓜(guā)的(dí)平均近紅外光(guāng)譜(pǔ) / Average near-infrared spectra of watermelon

使(shǐ)用LVF儀(yí)器(qì)預(yù)測(cè)完(wán)整(zhěng)條紋淺綠色(sè)和實(shí)心(xīn)深(shēn)綠色(sè)外(wài)皮西瓜中(zhōng)可(kě)溶性固形(xíng)物含(hán)量(%)的最佳(jiā)方(fāng)程的校準(zhǔn)統計(jì)量 /

Calibration statistics of the optimal equation for predicting soluble solids content (%) in intact striped light-green and solid dark-green rind watermelons using LVF instrumentation

榴蓮研(yán)究 / Durian Study：

(a)果柄(bǐng)和(b)果皮(pí)被放(fàng)置在樣品架中(zhōng)的(dí)情(qíng)況。通過旋轉(zhuǎn)旋鈕(niǔ)可(kě)水平(píng)或垂(chuí)直移動樣(yàng)品(pǐn)，如指針所示(shì)，使(shǐ)其(qí)到達檢測焦點位置(zhì)。

Photographs showing (a) the stem and (b) the rind placed in the sample holder. Knob rotations are used  to move the samples horizontally and vertically to the focal position for irradiation as indicated by the needle.

基(jī)於近紅(hóng)外光譜的(dí)不同成熟階段均值(zhí)光譜變化：(a) 果肉(ròu)；(b) 果皮；(c) 果柄

Variationwithrespect to maturation stages of mean spectra using near-infrared spectroscopy of: (a) pulp;  (b) rind; and (c) stem.

討論和結(jié)語 / Disscusion & Conclusion

實(shí)現(xiàn)穩(wěn)健(jiàn)的校(xiào)準模型是當(dāng)前(qián)研究的重(zhòng)點。模型的低穩健(jiàn)性會阻(zǔ)礙其(qí)在跨(kuà)環境（從實驗室到(dào)現場）
，跨(kuà)樣本（不(bù)同品(pǐn)種/年(nián)份(fèn)）以(yǐ)及(jí)跨設備間的(dí)推(tuī)廣應(yīng)用。自然界(jiè)的(dí)復雜(zá)性(xìng)和(hé)大(dà)量(liáng)變異是(shì)主(zhǔ)要挑戰。可構(gòu)建(jiàn)覆(fù)蓋不(bù)同(tóng)年份(fèn)
，果園和品種的(dí)多樣化樣品(pǐn)數據庫(kù)增(zēng)強模型適(shì)應(yīng)性。

模(mó)型(xíng)泛化能(néng)力研究(jiū)仍(réng)顯不(bù)足，實際(jì)生(shēng)產中的(dí)環境條件波動(dòng)也(yě)會進一步(bù)考(kǎo)驗模(mó)型普適(shì)性(xìng)。

作為高光譜成像係統硬件的(dí)提供(gōng)商，我(wǒ)們致力(lì)於為高校研(yán)究(jiū)所和(hé)解(jiě)決方(fāng)案(àn)集(jí)成(chéng)商提(tí)供高性能
，可靠的(dí)光(guāng)學成像(xiàng)平台(tái)。我(wǒ)們(mén)期(qī)待與研究(jiū)機構
，係(xì)統集成商(shāng)合作，共(gòng)同(tóng)開發麵向特(tè)定場(cháng)景的(dí)成熟度(dù)檢(jiǎn)測解決(jué)方(fāng)案(àn)，推(tuī)動(dòng)這(zhè)項(xiàng)技(jì)術(shù)從(cóng)實驗(yàn)室走向田(tián)間(jiān)和(hé)生產(chǎn)綫(xiàn)。

Developing robust calibration models remains a key focus of current research. Low model robustness hinders their application across environments (from lab to field), samples (different varieties/years), and devices. The complexity and vast variability in nature pose major challenges. Building diverse sample databases covering multiple years, orchards, and varieties can enhance model adaptability.

Research on model generalization capability is still insufficient, and fluctuating environmental conditions in real-world production further test model universality.

As a provider of hyperspectral imaging system hardware, we are committed to delivering high-performance, reliable optical imaging platforms to academic institutions and solution integrators. We look forward to collaborating with research organizations and system integrators to develop ripeness detection solutions tailored to specific scenarios, advancing this technology from the lab to fields and production lines.

案(àn)例來(lái)源(yuán) / Source：

1. Varga LA, Makowski J, Zell A. Measuring the ripeness of fruit with hyperspectral imaging and deep learning. 2021 International Joint Conference on Neural Networks (IJCNN). 2021:1-8.

2. Vega-Castellote M, Sánchez MT, Torres I, de la Haba MJ, Pérez-Marín D. Assessment of watermelon maturity using portable new generation NIR spectrophotometers. Scientia Horticulturae. 2022;304:111328. 

3. Somton W, Pathaveerat S, Terdwongworakul A. Application of near infrared spectroscopy for indirect evaluation of “Monthong" durian maturity. International Journal of Food Properties. 2015;18(6):1155-1168. 

4. Liu J, Meng H. Research on the maturity detection method of Korla pears based on hyperspectral technology. Agriculture. 2024;14:1257.