Figure - available from: Food and Bioprocess Technology
This content is subject to copyright. Terms and conditions apply.
VIS-NIR in-line design composed of camera, illumination, and linear actuator with reflection probe, where d0 is the range of the movement the probe, d1 is the distance between the centre of the rollers and the resting place of the probe, d2 is the known distance between the centre of the rollers and the surface of the fruit given by the image analysis system, and d3 is the distance measuring distance
Source publication
One of the most studied techniques for the non-destructive determination of the internal quality of fruits has been visible and near-infrared (VIS-NIR) reflectance spectroscopy. This work evaluates a new non-destructive in-line VIS-NIR spectroscopy prototype for in-line identification of five apple varieties, with the advantage that it allows the s...
Citations
... Traditional destructive techniques to assess fruit quality are timeconsuming, require complex equipment and sample preparation, and skilled operators need to perform the analyses. On the contrary, optical non-destructive alternatives for fruit quality monitoring, such as NIRS, have been proven faster, easier, and well-suited for in-line grading and sorting processes (Arendse et al., 2018;Cortés et al., 2019a). Spectroscopy measures how light interacts with a sample. ...
The lemon industry has the challenge of providing fruits with high-quality standards worldwide. Replacing the subjective fruit quality assessment methods with objective and non-destructive techniques. Total soluble solids (TSS) and titratable acidity (TA) have been revealed as important ripening markers in lemons. Therefore, this study proposes, for the first time, using near-infra-red spectroscopy (NIRS) as a rapid and non-destructive alternative to evaluate these quality traits in 'Fino' lemons (Citrus limon L. Burm) during ripeness. NIR spectra (950-1700 nm) of intact lemons collected from two different orchards at three ripening stages were acquired, while standard destructive methods were used to determine TSS and TA in the juice of each fruit. The prediction of the quality parameters was carried out using partial least squares regression (PLS-R) models. Three approaches were followed to validate the models: internal, external, and recalibrated external validation. The results following the first approach presented a good predictive performance for both quality parameters (TSS: R 2 = 0.84, RMSEP = 0.42 and RPD = 2.5; TA: R 2 = 0.72, RMSEP = 0.45 and RPD = 2.0). When the external validation was performed, the best results were obtained for the TSS prediction using recalibrated models, maintaining good predictive performance accuracy (R 2 = 0.74 and 0.67, RMSEP = 0.42 and 0.58, and RPD = 2.4 and 1.7). Regarding distinguishing different origins, models based on partial least squares discriminant analysis (PLS-DA) were externally validated, achieving 66.4% correct classification, respectively. Thus, applying NIR technology in the lemon fruit packinghouses is a promising alternative to improve fruit management and meet consumer demands.
... Micro-spectrometers have been widely used in micro-nano satellites [1], online detection [2,3], food safety [4,5], and other fields due to their advantages of a small size, easy integration, and good performance in special environments. Traditional dispersive spectrometers are mainly based on gratings, prisms, and other wavelength division methods [6,7]. ...
Micro-spectrometers have great potential in various fields such as medicine, agriculture, and aerospace. In this work, a quantum-dot (QD) light-chip micro-spectrometer is proposed in which QDs emit different wavelengths of light that are combined with a spectral reconstruction (SR) algorithm. The QD array itself can play the roles of both the light source and the wavelength division structure. The spectra of samples can be obtained by using this simple light source with a detector and algorithm, and the spectral resolution reaches 9.7 nm in the wavelength range from 580 nm to 720 nm. The area of the QD light chip is 4 × 7.5 mm ² , which is 20 times smaller than the halogen light sources of commercial spectrometers. It does not need a wavelength division structure and greatly reduces the volume of the spectrometer. Such a micro-spectrometer can be used for material identification: in a demonstration, three kinds of transparent samples, real and fake leaves, and real and fake blood were classified with an accuracy of 100%. These results indicate that the spectrometer based on a QD light chip has broad application prospects.
... In the context of apple traceability, NIR combined with chemometrics has been successfully used to classify apples based on their origin, cultivar, and quality. It is possible to develop predictive models for the classification of apples by Vis/NIR spectroscopy according to their origin and cultivar, forming a highly practical and flexible method for the traceability of apples [279][280][281]. For example, Eisenstecken et al. demonstrated that high accuracy rates in classifying apples according to their cultivar and orchard elevation can be delivered by spectroscopic methods in the conventional NIR region [279]. ...
Spectroscopic methods deliver a valuable non-destructive analytical tool that provides simultaneous qualitative and quantitative characterization of various samples. Apples belong to the world's most consumed crops and with the current challenges of climate change and human impacts on the environment, maintaining high-quality apple production has become critical. This review comprehensively analyzes the application of spectroscopy in near-infrared (NIR) and visible (Vis) regions, which not only show particular potential in evaluating the quality parameters of apples but also in optimizing their production and supply routines. This includes the assessment of the external and internal characteristics such as color, size, shape, surface defects, soluble solids content (SSC), total titratable acidity (TA), firmness, starch pattern index (SPI), total dry matter concentration (DM), and nutritional value. The review also summarizes various techniques and approaches used in Vis/NIR studies of apples, such as authenticity, origin, identification, adulteration, and quality control. Optical sensors and associated methods offer a wide suite of solutions readily addressing the main needs of the industry in practical routines as well, e.g., efficient sorting and grading of apples based on sweetness and other quality parameters, facilitating quality control throughout the production and supply chain. This review also evaluates ongoing development trends in the application of handheld and portable instruments operating in the Vis/NIR and NIR spectral regions for apple quality control. The use of these technologies can enhance apple crop quality, maintain competitiveness, and meet the demands of consumers, making them a crucial topic in the apple industry. The focal point of this review is placed on the literature published in the last five years, with the exceptions of seminal works that have played a critical role in shaping the field or representative studies that highlight the progress made in specific areas.
... Fresh apples are one of the main fruits consumed worldwide due to their nutritional value and health benefits, with more than 7500 varieties available, the most popular ones including Granny Smith, Gala, Fuji, Red Delicious, Golden Delicious, and Braeburn [1,2]. Typically, several varieties of apples are grown simultaneously in an apple orchard, so it is easy to mix different varieties of apples with similar appearances during the harvesting and marketing process [3]. Although different varieties of apples are similar in external appearance, they have different intrinsic qualities in terms of taste and nutritional value, which is why people spend a lot of time sorting, packing, and labeling apples before selling them [4,5]. ...
In this study, series networks (AlexNet and VGG-19) and directed acyclic graph (DAG) networks (ResNet-18, ResNet-50, and ResNet-101) with transfer learning were employed to identify and classify 13 classes of apples from 7439 images. Two training datasets, model evaluation metrics, and three visualization methods were used to objectively assess, compare, and interpret five Convolutional Neural Network (CNN)-based models. The results show that the dataset configuration had a significant impact on the classification results, as all models achieved over 96.1% accuracy on dataset A (training-to-testing = 2.4:1.0) compared to 89.4–93.9% accuracy on dataset B (training-to-testing = 1.0:3.7). VGG-19 achieved the highest accuracy of 100.0% on dataset A and 93.9% on dataset B. Moreover, for networks of the same framework, the model size, accuracy, and training and testing times increased as the model depth (number of layers) increased. Furthermore, feature visualization, strongest activations, and local interpretable model-agnostic explanations techniques were used to show the understanding of apple images by different trained models, as well as to reveal how and why the models make classification decisions. These results improve the interpretability and credibility of CNN-based models, which provides guidance for future applications of deep learning methods in agriculture.
... The heterogeneous nature of the sample is accounted for by techniques such as taking three or four different types of measurements; by the use of 120 and 90-degree rotations (54,58); and by the use of fabricated fruit holders (59,61) or by using the arrangement in the spectrometer. Standard NIR models require whole fruit cover scanning using arrangements of diode-array instruments (44), the integration of the sphere around them to recover information that is otherwise lost (52), and methods to keep the distance from fruit to measurement probe constant, irrespective of the size (50). ...
... Because fluctuations in temperature and light play a vital role in creating the NIR model, care should be taken to keep the surrounding environment the same for all measurements (53). For cultivar prediction, the input data are preprocessed using methods such as multiplicative scatter correction (MSC) (51,55,60), EMSC (50,62), standard normal variate (SNV) (7,51,56), detrend (51,55), normalization (58,59), Savitzky-Golay (7,50,51,61), and Norris gap (55). These are independent reference techniques for eliminating the unwanted effects of irrelevant information in the spectra (52). ...
... Because fluctuations in temperature and light play a vital role in creating the NIR model, care should be taken to keep the surrounding environment the same for all measurements (53). For cultivar prediction, the input data are preprocessed using methods such as multiplicative scatter correction (MSC) (51,55,60), EMSC (50,62), standard normal variate (SNV) (7,51,56), detrend (51,55), normalization (58,59), Savitzky-Golay (7,50,51,61), and Norris gap (55). These are independent reference techniques for eliminating the unwanted effects of irrelevant information in the spectra (52). ...
The qualitative and quantitative evaluation of agricultural products has often been carried out using traditional, i.e., destructive, techniques. Due to their inherent disadvantages, non-destructive methods that use near-infrared spectroscopy (NIRS) coupled with chemometrics could be useful for evaluating various agricultural products. Advancements in computational power, machine learning, regression models, artificial neural networks (ANN), and other predictive tools have made their way into NIRS, improving its potential to be a feasible alternative to destructive measurements. Moreover, the incorporation of suitable preprocessing techniques and wavelength selection methods has arguably proven its practical feasibility. This review focuses on the various computation methods used for processing the spectral data collected and discusses the potential applications of NIRS for evaluating the quality and safety of agricultural products. The challenges associated with this technology are also discussed, as well as potential future perspectives. We conclude that NIRS is a potentially useful tool for the rapid assessment of the quality and safety of agricultural products.
... Its structural schematic diagram is shown in Figure 1a. The speed of the conveyor was set as 8 cm/s according to the preliminary research results of our laboratory and the existing reports on online detection [28][29][30]. The halogen lamps were installed on both sides of the dark box, and the installation angle was set to 30° to the horizontal plane. ...
Single-probe near-infrared spectroscopy (NIRS) usually uses different spectral information for modelling, but there are few reports about its influence on model performance. Based on sized-adaptive online NIRS information and the 2D conventional neural network (CNN), minced samples of pure mutton, pork, duck, and adulterated mutton with pork/duck were classified in this study. The influence of spectral information, convolution kernel sizes, and classifiers on model performance was separately explored. The results showed that spectral information had a great influence on model accuracy, of which the maximum difference could reach up to 12.06% for the same validation set. The convolution kernel sizes and classifiers had little effect on model accuracy but had significant influence on classification speed. For all datasets, the accuracy of the CNN model with mean spectral information per direction, extreme learning machine (ELM) classifier, and 7 × 7 convolution kernel was higher than 99.56%. Considering the rapidity and practicality, this study provides a fast and accurate method for online classification of adulterated mutton.
... In the last years, the use of portable devices grew with its development and marketing (Pasquini, 2018). The miniaturization has a great impact in the industrial environment as it makes it more feasible to apply NIR technology in for in-line routine monitoring along the production chain; additionally (Cortés et al., 2019;González-Martín et al., 2021;. Table 1 summarizes the main characteristics for each type of device, based on manufacturer's specifications, reported applications and our expertise. ...
Wheat flour is a food ingredient used in different processed food products, including pasta, cake, bread, among others. Therefore, the authentication and assurance of good quality are of great importance. The traditional techniques used for quality parameters determinations are laborious, destructive, and demand chemical analysis. Hence, it is necessary the development of techniques capable to overcome these disadvantages. The spectral techniques are rapid, non-destructive, and chemical-free. This review approaches the applications of Near Infrared (NIR) Spectroscopy, Fourier Transform Near-Infrared (FT-NIR) Spectroscopy, and Hyperspectral Imaging (HSI) in the wheat flour and wheat-based products authentication and assessment of quality parameters, composition, and contamination. Considering the need from the processing industry for a rapid analysis, moving the bench-top analytical system to the production line, future studies can explore the in/on-line applications of these techniques for industrial process lines and compare the use of handheld and benchtop spectrometers in these applications.
... A Partial Least Squares (PLS) model is proposed here to relate FTIR measurements and rheological model parameters. Chemometric approaches are recently been used to relate food properties and measurements obtained by different spectroscopic techniques (Carbas et al., 2020;Cortés et al., 2019;Cueto et al., 2018;Liu et al., 2014a, b;Wu et al., 2014). ...
Rheological measurements and FTIR spectroscopy were used to characterize different doughs, obtained by commercial and monovarietal durum wheat flours (Cappelli and Karalis). Rheological frequency sweep tests were carried out, and the Weak Gel model, whose parameters may be related to gluten network extension and strength, was applied. IR analysis mainly focused on the Amide III band, revealing significant variations in the gluten network. Compared to the other varieties, Karalis semolina showed a higher amount of α-helices and a lower amount of β-sheets and random structures. Spectroscopic and rheological data were then correlated using Partial Least Squares regression (PLS) coupled with the Variable Importance in Projection (VIP) technique. The combined use of the techniques provided useful insights into the interplay among protein structures, gluten network features, and rheological properties. In detail, β-sheets and α-helices protein conformations were shown to significantly affect the gluten network's mechanical strength.
... These methods have achieved relevant detection results; however, the speed and visibility of these methods are partially limited. In addition, although X-ray (Guelpa et al., 2015), magnetic resonance imaging (MRI) (Ezeanaka et al., 2019), visible spectrum (Cortés et al., 2019), electronic nose (Ezhilan et al., 2020), and near-infrared spectrum (Shahin et al., 2014) have great potential in assessing fruit bruising, such equipment is complex, expensive, and not suitable for conventional bruise testing. And compared to above methods, thermal imaging technology with cheap and simple equipment, is applicable for more complex environments to detection because of its properties of no need for a particular condition (Zeng et al., 2020). ...
The detection of bruises plays a vital role in the quality evaluation of strawberries. This study aimed to detect strawberry bruises based on thermal images and classify bruises using a convolutional neural network (CNN). A simple active thermal imaging system was used to capture 2903 thermal images collected from 400 strawberries over 5 days. Moreover, the temperature difference between the bruised area and the unbruised area of the strawberry over time was analyzed. Some of the most advanced pretrained CNN models and the optimized CNN model were evaluated for the classification of unbruised and bruised strawberries based on collected thermal images. The results show that the accuracy of the optimized CNN network is 0.98, which is much higher than the accuracy of the pretrained models. Thus, this study provides a high degree of accuracy in the classification of unbruised and bruised strawberries using the optimized CNN model based on its thermal images, indicating which can be an effective method of detecting and classifying strawberries.
... For this reason, PCA algorithms search for the maximum variance direction, in the multidimensional space of the original dataset (Fig. 19). VIS-NIR or NIR spectroscopic techniques coupled with PCA show a good ability to detect a large apple variability due to varieties (Cortés et al., 2019;Daniela Eisenstecken et al., 2019;He et al., 2005 and geographical origins (Daniela Eisenstecken et al., 2019;Schmutzler et al., 2014), as well as the concentrations of different mixed fruit purees (Contal et al., 2002). ...
This thesis aimed to show how vibrational spectroscopy including near infrared (NIR), mid infrared (MIR), Raman and NIR hyperspectral imaging (NIR-HSI) coupled with advanced chemometrics can highlight the variability and heterogeneity of both, raw apples and processed purees. Experimental trials were designed to modulate several factors in orchard (varieties, agricultural practices), during post-harvest storage (4°C) and processing (temperature, grinding and refining) in order to modify properties and composition of apples and purees. An efficient approach using NIR-HSI allowed illustrating the distribution of total sugars and dry matter inside apples. The inter-batch variability of apples and the intra-batch variability between individual apples intensively changed the cooked purees. MIR spectroscopy was the best tool to detect the variability of purees and assess their biochemical (soluble solids, acidity, dry matter, fructose, sucrose and malic acid), rheological (viscosity and viscoelastic moduli) and textural (particle size and volume) properties. Good linear correlations were found between apple texture and puree viscosity, as well as between apple and puree acidity, soluble solids and dry matter. Therefore, VIS-NIR and NIR techniques allowed to predict the taste and texture of purees from the non-destructive spectra of apples. Besides, MIR spectroscopy can guide puree formulation from spectra of single-variety purees. Our innovative approaches could provide objective data to better manage apples and to adapt processing conditions according to their initial properties. The ultimate goal is to improve the quality of fresh and processed fruits while reducing losses.
