How to predict the sugariness and hardness of melons: A near-infrared hyperspectral imaging method

Melon ripeness detection by an improved object detection algorithm for resource constrained environments

BACKGROUND

Ripeness is a phenotype that significantly impacts the quality of fruits, constituting a crucial factor in the cultivation and harvesting processes. Manual detection methods and experimental analysis, however, are inefficient and costly.

RESULTS

In this study, we propose a lightweight and efficient melon ripeness detection method, MRD-YOLO, based on an improved object detection algorithm. The method combines a lightweight backbone network, MobileNetV3, a design paradigm Slim-neck, and a Coordinate Attention mechanism. Additionally, we have created a large-scale melon dataset sourced from a greenhouse based on ripeness. This dataset contains common complexities encountered in the field environment, such as occlusions, overlapping, and varying light intensities. MRD-YOLO achieves a mean Average Precision of 97.4% on this dataset, achieving accurate and reliable melon ripeness detection. Moreover, the method demands only 4.8 G FLOPs and 2.06 M parameters, representing 58.5% and 68.4% of the baseline YOLOv8n model, respectively. It comprehensively outperforms existing methods in terms of balanced accuracy and computational efficiency. Furthermore, it maintains real-time inference capability in GPU environments and demonstrates exceptional inference speed in CPU environments. The lightweight design of MRD-YOLO is anticipated to be deployed in various resource constrained mobile and edge devices, such as picking robots. Particularly noteworthy is its performance when tested on two melon datasets obtained from the Roboflow platform, achieving a mean Average Precision of 85.9%. This underscores its excellent generalization ability on untrained data.

CONCLUSIONS

This study presents an efficient method for melon ripeness detection, and the dataset utilized in this study, alongside the detection method, will provide a valuable reference for ripeness detection across various types of fruits.

A Performance Evaluation of Two Hyperspectral Imaging Systems for the Prediction of Strawberries' Pomological Traits

Pomological traits are the major factors determining the quality and price of fresh fruits. This research was aimed to investigate the feasibility of using two hyperspectral imaging (HSI) systems in the wavelength regions comprising visible to near infrared (VisNIR) (400-1000 nm) and short-wave infrared (SWIR) (935-1720 nm) for predicting four strawberry quality attributes (firmness-FF, total soluble solid content-TSS, titratable acidity-TA, and dry matter-DM). Prediction models were developed based on artificial neural networks (ANN). The entire strawberry VisNIR reflectance spectra resulted in accurate predictions of TSS (R2 = 0.959), DM (R2 = 0.947), and TA (R2 = 0.877), whereas good prediction was observed for FF (R2 = 0.808). As for models from the SWIR system, good correlations were found between each of the physicochemical indices and the spectral information (R2 = 0.924 for DM; R2 = 0.898 for TSS; R2 = 0.953 for TA; R2 = 0.820 for FF). Finally, data fusion demonstrated a higher ability to predict fruit internal quality (R2 = 0.942 for DM; R2 = 0. 981 for TSS; R2 = 0.976 for TA; R2 = 0.951 for FF). The results confirmed the potential of these two HSI systems as a rapid and nondestructive tool for evaluating fruit quality and enhancing the product's marketability.

Nondestructive quality assessment and maturity classification of loquats based on hyperspectral imaging

The traditional method for assessing the quality and maturity of loquats has disadvantages such as destructive sampling and being time-consuming. In this study, hyperspectral imaging technology was used to nondestructively predict and visualise the colour, firmness, and soluble solids content (SSC) of loquats and discriminate maturity. On comparison of the performance of different feature variables selection methods and the calibration models, the results indicated that the multiple linear regression (MLR) models combined with the competitive adaptive reweighting algorithm (CARS) yielded the best prediction performance for loquat quality. Particularly, CARS-MLR models with optimal prediction performance were obtained for the colour (R2P = 0.96, RMSEP = 0.45, RPD = 5.38), firmness (R2P = 0.87, RMSEP = 0.23, RPD = 2.81), and SSC (R2P = 0.84, RMSEP = 0.51, RPD = 2.54). Subsequently, distribution maps of the colour, firmness, and SSC of loquats were obtained based on the optimal CARS-MLR models combined with pseudo-colour technology. Finally, on comparison of different classification models for loquat maturity, the partial least square discrimination analysis model demonstrated the best performance, with classification accuracies of 98.19% and 97.99% for calibration and prediction sets, respectively. This study demonstrated that the hyperspectral imaging technique is promising for loquat quality assessment and maturity classification.

Application of Hyperspectral Imaging as a Nondestructive Technology for Identifying Tomato Maturity and Quantitatively Predicting Lycopene Content

Maturity is a crucial indicator in assessing the quality of tomatoes, and it is closely related to lycopene content. Using hyperspectral imaging, this study aimed to monitor tomato maturity and predict its lycopene content at different maturity stages. Standard normal variable (SNV) transformation was applied to preprocess the hyperspectral data. Then, using competitive adaptive reweighted sampling (CARS), the characteristic wavelengths were selected to simplify the calibration models. Based on the full and characteristic wavelengths, a support vector classifier (SVC) model was developed to determine tomato maturity qualitatively. The results demonstrated that the classification accuracy using the characteristic wavelength led to the obtention of better results with an accuracy of 95.83%. In addition, the support vector regression (SVR) and partial least squares regression (PLSR) models were utilized to predict lycopene content. With a coefficient of determination for prediction (R2P) of 0.9652 and a root mean square error for prediction (RMSEP) of 0.0166 mg/kg, the SVR model exhibited the best quantitative prediction capacity based on the characteristic wavelengths. Following this, a visual distribution map was created to evaluate the lycopene content in tomato fruit intuitively. The results demonstrated the viability of hyperspectral imaging for detecting tomato maturity and quantitatively predicting the lycopene content during storage.

Current Application of Advancing Spectroscopy Techniques in Food Analysis: Data Handling with Chemometric Approaches

In today's era of increased food consumption, consumers have become more demanding in terms of safety and the quality of products they consume. As a result, food authorities are closely monitoring the food industry to ensure that products meet the required standards of quality. The analysis of food properties encompasses various aspects, including chemical and physical descriptions, sensory assessments, authenticity, traceability, processing, crop production, storage conditions, and microbial and contaminant levels. Traditionally, the analysis of food properties has relied on conventional analytical techniques. However, these methods often involve destructive processes, which are laborious, time-consuming, expensive, and environmentally harmful. In contrast, advanced spectroscopic techniques offer a promising alternative. Spectroscopic methods such as hyperspectral and multispectral imaging, NMR, Raman, IR, UV, visible, fluorescence, and X-ray-based methods provide rapid, non-destructive, cost-effective, and environmentally friendly means of food analysis. Nevertheless, interpreting spectroscopy data, whether in the form of signals (fingerprints) or images, can be complex without the assistance of statistical and innovative chemometric approaches. These approaches involve various steps such as pre-processing, exploratory analysis, variable selection, regression, classification, and data integration. They are essential for extracting relevant information and effectively handling the complexity of spectroscopic data. This review aims to address, discuss, and examine recent studies on advanced spectroscopic techniques and chemometric tools in the context of food product applications and analysis trends. Furthermore, it focuses on the practical aspects of spectral data handling, model construction, data interpretation, and the general utilization of statistical and chemometric methods for both qualitative and quantitative analysis. By exploring the advancements in spectroscopic techniques and their integration with chemometric tools, this review provides valuable insights into the potential applications and future directions of these analytical approaches in the food industry. It emphasizes the importance of efficient data handling, model development, and practical implementation of statistical and chemometric methods in the field of food analysis.

"How sweet are your strawberries?": Predicting sugariness using non-destructive and affordable hardware

Global soft fruit supply chains rely on trustworthy descriptions of product quality. However, crucial criteria such as sweetness and firmness cannot be accurately established without destroying the fruit. Since traditional alternatives are subjective assessments by human experts, it is desirable to obtain quality estimations in a consistent and non-destructive manner. The majority of research on fruit quality measurements analyzed fruits in the lab with uniform data collection. However, it is laborious and expensive to scale up to the level of the whole yield. The "harvest-first, analysis-second" method also comes too late to decide to adjust harvesting schedules. In this research, we validated our hypothesis of using in-field data acquirable via commodity hardware to obtain acceptable accuracies. The primary instance that the research concerns is the sugariness of strawberries, described by the juice's total soluble solid (TSS) content (unit: °Brix or Brix). We benchmarked the accuracy of strawberry Brix prediction using convolutional neural networks (CNN), variational autoencoders (VAE), principal component analysis (PCA), kernelized ridge regression (KRR), support vector regression (SVR), and multilayer perceptron (MLP), based on fusions of image data, environmental records, and plant load information, etc. Our results suggest that: (i) models trained by environment and plant load data can perform reliable prediction of aggregated Brix values, with the lowest RMSE at 0.59; (ii) using image data can further supplement the Brix predictions of individual fruits from (i), from 1.27 to as low up to 1.10, but they by themselves are not sufficiently reliable.

Visualization of Sugar Content Distribution of White Strawberry by Near-Infrared Hyperspectral Imaging

In this study, an approach to visualize the spatial distribution of sugar content in white strawberry fruit flesh using near-infrared hyperspectral imaging (NIR-HSI; 913-2166 nm) is developed. NIR-HSI data collected from 180 samples of "Tochigi iW1 go" white strawberries are investigated. In order to recognize the pixels corresponding to the flesh and achene on the surface of the strawberries, principal component analysis (PCA) and image processing are conducted after smoothing and standard normal variate (SNV) pretreatment of the data. Explanatory partial least squares regression (PLSR) analysis is performed to develop an appropriate model to predict Brix reference values. The PLSR model constructed from the raw spectra extracted from the flesh region of interest yields high prediction accuracy with an RMSEP and R2p values of 0.576 and 0.841, respectively, and with a relatively low number of PLS factors. The Brix heatmap images and violin plots for each sample exhibit characteristics feature of sugar content distribution in the flesh of the strawberries. These findings offer insights into the feasibility of designing a noncontact system to monitor the quality of white strawberries.

Quantification of Photosynthetic Pigments in Neopyropia yezoensis Using Hyperspectral Imagery

Phycobilisomes and chlorophyll-a (Chla) play important roles in the photosynthetic physiology of red macroalgae and serve as the primary light-harvesting antennae and reaction center for photosystem II. Neopyropia is an economically important red macroalga widely cultivated in East Asian countries. The contents and ratios of 3 main phycobiliproteins and Chla are visible traits to evaluate its commercial quality. The traditional analytical methods used for measuring these components have several limitations. Therefore, a high-throughput, nondestructive, optical method based on hyperspectral imaging technology was developed for phenotyping the pigments phycoerythrin (PE), phycocyanin (PC), allophycocyanin (APC), and Chla in Neopyropia thalli in this study. The average spectra from the region of interest were collected at wavelengths ranging from 400 to 1000 nm using a hyperspectral camera. Following different preprocessing methods, 2 machine learning methods, partial least squares regression (PLSR) and support vector machine regression (SVR), were performed to establish the best prediction models for PE, PC, APC, and Chla contents. The prediction results showed that the PLSR model performed the best for PE (R _Test ² = 0.96, MAPE = 8.31%, RPD = 5.21) and the SVR model performed the best for PC (R _Test ² = 0.94, MAPE = 7.18%, RPD = 4.16) and APC (R _Test ² = 0.84, MAPE = 18.25%, RPD = 2.53). Two models (PLSR and SVR) performed almost the same for Chla (PLSR: R _Test ² = 0.92, MAPE = 12.77%, RPD = 3.61; SVR: R _Test ² = 0.93, MAPE = 13.51%, RPD =3.60). Further validation of the optimal models was performed using field-collected samples, and the result demonstrated satisfactory robustness and accuracy. The distribution of PE, PC, APC, and Chla contents within a thallus was visualized according to the optimal prediction models. The results showed that hyperspectral imaging technology was effective for fast, accurate, and noninvasive phenotyping of the PE, PC, APC, and Chla contents of Neopyropia in situ. This could benefit the efficiency of macroalgae breeding, phenomics research, and other related applications.

A non-destructive methodology for determination of cantaloupe sugar content using machine vision and deep learning

BACKGROUND

To determine the maturity of cantaloupe, measuring the soluble solid content (SSC) as the indicator of sugar content based on the refractometric index is commonly practised. This method, however, is destructive and limited to a small number of samples. In this study, the coupling of a convolutional neural network (CNN) with machine vision was proposed in detecting the SSC of cantaloupe. The cantaloupe images were first acquired under controlled and uncontrolled conditions and subsequently fed to the CNN to predict the class to which each cantaloupe belonged. Four hand-crafted classical machine-learning classifiers were used to compare against the performance of the CNN.

RESULTS

Experimental results showed that the CNN method significantly outperformed others, with an improvement of >100% being achieved in terms of classification accuracy, considering the data acquired under the uncontrolled environment.

CONCLUSION

The results demonstrated the potential benefit to operationalize CNNs in practice for SSC determination of cantaloupe before harvesting. © 2022 Society of Chemical Industry.

Differentiation of Organic Cocoa Beans and Conventional Ones by Using Handheld NIR Spectroscopy and Multivariate Classification Techniques

The global market for organic cocoa beans continues to show sturdy growth. A low-cost handheld NIR spectrometer (900-1700 nm) combined with multivariate classification algorithms was used for rapid differentiation analysis of organic cocoa beans' integrity. In this research, organic and conventionally cultivated cocoa beans were collected from different locations in Ghana and scanned nondestructively with a handheld spectrometer. Different preprocessing treatments were employed. Principal component analysis (PCA) and classification analysis, RF (random forest), KNN (K-nearest neighbours), LDA (linear discriminant analysis), and PLS-DA (partial least squares-discriminant analysis) were performed comparatively to build classification models. The performance of the models was evaluated by accuracy, specificity, sensitivity, and efficiency. Second derivative preprocessing together with PLS-DA algorithm was superior to the rest of the algorithms with a classification accuracy of 100.00% in both the calibration set and prediction set. Second derivative algorithm was found to be the best preprocessing tool. The identification rates for the calibration set and prediction set were 96.15% and 98.08%, respectively, for RF, 91.35% and 92.31% for KNN, and 90.38% and 98.08% for LDA. Generally, the results showed that a handheld NIR spectrometer coupled with an appropriate multivariate algorithm could be used in situ for the differentiation of organic cocoa beans from conventional ones to ensure food integrity along the cocoa bean value chain.

Modeling relationship between protein oxidation and denaturation and texture, moisture loss of bighead carp (Aristichthys Nobilis) during frozen storage

To evaluate the effects of protein oxidation and denaturation on the fish texture and moisture loss during frozen storage, we measured the changes of protein oxidation and denaturation (salt-soluble protein (SSP), total sulfhydryl (SH), disulfide (SS), carbonyl contents and Ca²⁺ -ATPase activity), texture (hardness), and moisture loss (drip loss) of bighead carp fillets stored at -12, -20 and -28°C during 16 weeks. These data were employed to develop partial least squares regression (PLSR) model, radial basis function neural network (RBFNN) model, PLSR-RBFNN (PR) model and RBFNN-PLSR (RP) model. The results showed that the RP model provided no enhancement to RBFNN model because it had the exactly same root mean square error (RMSE) and R² . PLSR model showed better performance than other models when predicting hardness. More appropriate linear or linearity-dominant hybrid model needed to be explored to establish the relationship between protein oxidation and denaturation and texture. The PR model performed better than other models in predicting drip loss with its lower RMSE and higher R² , which revealed both linear and nonlinear relationship between protein oxidation and denaturation and moisture loss. Therefore, the PR model was a promising and encouraging tool to provide a more comprehensive understanding of the relationship between protein oxidation and denaturation and moisture loss of fish during frozen storage. PRACTICAL APPLICATION: The study explored the effects of protein oxidation and denaturation on the texture and moisture loss of bighead carp during frozen storage (-12 to -28°C). PLSR model showed better performance than other models when predicting the relationship between protein oxidation and denaturation and texture. The PR model was an available tool for manufacturers to predict the relationship between protein oxidation and denaturation and moisture loss.

Identifying Individual Nutrient Deficiencies of Grapevine Leaves Using Hyperspectral Imaging

The efficiency of a vineyard management system is directly related to the effective management of nutritional disorders, which significantly downgrades vine growth, crop yield and wine quality. To detect nutritional disorders, we successfully extracted a wide range of features using hyperspectral (HS) images to identify healthy and individual nutrient deficiencies of grapevine leaves. Features such as mean reflectance, mean first derivative reflectance, variation index, mean spectral ratio, normalised difference vegetation index (NDVI) and standard deviation (SD) were employed at various stages in the ultraviolet (UV), visible (VIS) and near-infrared (N.I.R.) regions for our experiment. Leaves were examined visually in the laboratory and grouped as either healthy (i.e. control) or unhealthy. Then, the features of the leaves were extracted from these two groups. In a second experiment, features of individual nutrient-deficient leaves (e.g., N, K and Mg) were also analysed and compared with those of control leaves. Furthermore, a customised support vector machine (SVM) was used to demonstrate that these features can be utilised with a high degree of effectiveness to identify unhealthy samples and not only to distinguish from control and nutrient deficient but also to identify individual nutrient defects. Therefore, the proposed work corroborated that HS imaging has excellent potential to analyse features based on healthiness and individual nutrient deficiencies of grapevine leaves.

Non-destructive sugar content assessment of multiple cultivars of melons by dielectric properties

BACKGROUND

Non-destructive determination of the internal quality of fruit with a thick rind and of a large size is always difficult and challenging. To investigate the feasibility of the dielectric spectroscopy technique with respect to determining the sugar content of melons during the postharvest stage, three cultivars of melon samples (160 melons for each cultivar) were used to acquire dielectric spectra over the frequency range 20-4500 MHz. The three cultivars of melons were divided separately into a calibration set and a prediction set in a ratio of 3:1 by a joint x-y distance algorithm. Partial least squares (PLS) and extreme learning machine (ELM) methods were applied to develop individual-cultivar and multi-cultivar models based on full frequencies (FFs) and effective dielectric frequencies (EDFs) selected by the successive projection algorithm (SPA).

RESULTS

The results showed that ELM models demonstrated a better performance than PLS models for the same input dielectric variables. Most of the models built based on the EDFs selected by SPA had a slightly worse performance compared to those based on FFs. For both PLS and ELM methods, the models for multi-cultivars demonstrated a worse calibration and prediction performance compared to those for individual cultivars. When individual-cultivar and multi-cultivar samples were used to build sugar content determination models, the best model was FFs-ELM (R_p  = 0.887, RMSEP = 0.986), FFs-ELM (R_p  = 0.870, RMSEP = 1.028), FFs-PLS (R_p  = 0.882, RMSEP = 1.010) and FFs-ELM (R_p  = 0.849, RMSEP = 1.085) for 'Hongyanliang', 'Xinzaomi', 'Manao' and multi-cultivar melons, respectively.

CONCLUSION

The present study indicates that it is possible to develop both individual-cultivar and multi-cultivar models for determining the sugar content of melons based on the dielectric spectroscopy technique. © 2021 Society of Chemical Industry.

Stability assessment of extracts obtained from Arbutus unedo L. fruits in powder and solution systems using machine-learning methodologies

Arbutus unedo L. (strawberry tree) has showed considerable content in phenolic compounds, especially flavan-3-ols (catechin, gallocatechin, among others). The interest of flavan-3-ols has increased due their bioactive actions, namely antioxidant and antimicrobial activities, and by association of their consumption to diverse health benefits including the prevention of obesity, cardiovascular diseases or cancer. These compounds, mainly catechin, have been showed potential for use as natural preservative in foodstuffs; however, their degradation is increased by pH and temperature of processing and storage, which can limit their use by food industry. To model the degradation kinetics of these compounds under different conditions of storage, three kinds of machine learning models were developed: i) random forest, ii) support vector machine and iii) artificial neural network. The selected models can be used to track the kinetics of the different compounds and properties under study without the prior knowledge requirement of the reaction system.

Nondestructive authentication of the regional and geographical origin of cocoa beans by using a handheld NIR spectrometer and multivariate algorithm

Traceability in the cocoa bean trade is vital to ensuring quality. In this study, a handheld near-infrared (NIR) spectrometer was attempted for rapid and nondestructive regional and geographical classification of cocoa beans from different locations. Cocoa bean samples collected from seven cocoa-producing regions in Ghana (Eastern, Ashanti, Volta, Western South, Western North, Central, and Brong Ahafo) and four cocoa-producing countries in Africa (Uganda, Ivory Coast, Nigeria, and Ghana) were used. Among the preprocessing techniques employed, multiplicative scatter correction (MSC) performed better. The correct classification rate for the seven cocoa-producing regions in Ghana was 100% for LDA and SVM models in the training set and testing set. For classification of cocoa beans based on the country of origin, LDA and SVM also gave 100% classification rate both in the training set and testing set. The results give strong indications that hand-held spectroscopy coupled with chemometrics could be employed to provide the quick, accurate, and nondestructive classification of cocoa beans according to different locations. This technique could improve the work of quality control inspectors both from industry and regulatory perspectives for effective and quick detection of cocoa bean fraud.

Non-Destructive Detection of Strawberry Quality Using Multi-Features of Hyperspectral Imaging and Multivariate Methods

Soluble solid content (SSC), pH, and vitamin C (VC) are considered as key parameters for strawberry quality. Spectral, color, and textural features from hyperspectral reflectance imaging of 400-1000 nm was to develop the non-destructive detection approaches for SSC, pH, and VC of strawberries by integrating various multivariate methods as partial least-squares regression (PLSR), support vector regression, and locally weighted regression (LWR). SSC, pH, and VC of 120 strawberries were statistically analyzed to facilitate the partitioning of data sets, which helped optimize the model. PLSR, with spectral and color features, obtained the optimal prediction of SSC with determination coefficient of prediction (Rp2) of 0.9370 and the root mean square error of prediction (RMSEP) of 0.1145. Through spectral features, the best prediction for pH was obtained by LWR with Rp2 = 0.8493 and RMSEP = 0.0501. Combination of spectral and textural features with PLSR provided the best results of VC with Rp2 = 0.8769 and RMSEP = 0.0279. Competitive adaptive reweighted sampling and uninformative variable elimination (UVE) were used to select important variables from the above features. Based on the important variables, the accuracy of SSC, pH, and VC prediction both gain the promotion. Finally, the distribution maps of SSC, pH, and VC over time were generated, and the change trend of three quality parameters was observed. Thus, the proposed method can nondestructively and accurately determine SSC, pH, and VC of strawberries and is expected to design and construct the simple sensors for the above quality parameters of strawberries.

Standard Analytical Methods, Sensory Evaluation, NIRS and Electronic Tongue for Sensing Taste Attributes of Different Melon Varieties

Grafting by vegetables is a practice with many benefits, but also with some unknown influences on the chemical composition of the fruits. Our goal was to assess the effects of grafting and storage on the extracted juice of four orange-fleshed Cantaloupe type (Celestial, Donatello, Centro, Jannet) melons and two green-fleshed Galia types (Aikido, London), using sensory profile analysis and analytical instruments: An electronic tongue (E-tongue) and near-infrared spectroscopy (NIRS). Both instruments are known for rapid qualitative and quantitative food analysis. Linear discriminant analysis (LDA) was used to classify melons according to their varieties and storage conditions. Partial least square regression (PLSR) was used to predict sensory and standard analytical parameters. Celestial variety had the highest intensity for sensory attributes in Cantaloupe variety. Both green and orange-fleshed melons were discriminated and predicted in LDA with high accuracies (100%) using the E-tongue and NIRS. Galia and Cantaloupe inter-varietal classification with the E-tongue was 89.9% and 82.33%, respectively. NIRS inter-varietal classification was 100% with Celestial variety being the most discriminated as with the sensory results. Both instruments, classified different storage conditions of melons (grafted and self-rooted) with high accuracies. PLSR showed high accuracy for some standard analytical parameters, where significant differences were found comparing different varieties in ANOVA.

Rapid detection of cadmium and its distribution in Miscanthus sacchariflorus based on visible and near-infrared hyperspectral imaging

Monitoring the effectiveness of Miscanthus sacchariflorus to meet the basic requirements for environmental remediation projects is an important step in determining its use as a productive bioenergy crop for phytoremediation. Conventional chemical methods for the determination of cadmium (Cd) contents involve time-consuming, monotonous and destructive procedures and are not suitable for high-throughput screening. In the present study, visible and near-infrared hyperspectral imaging technology combined with chemometric methods was used to assess the Cd concentrations in M. sacchariflorus. The total Cd concentrations in different plant tissues were measured using an inductively coupled plasma-mass spectrometer. Partial least-squares regression and least-squares support vector machine were implemented to estimate Cd contents from spectral reflectance. Successive projections algorithm and competitive adaptive reweighted sampling (CARS) methodology were used for selecting optimal wavelength. The CARS-partial least-squares regression model resulted in the most accurate predictions of Cd contents in M. sacchariflorus leaves, with a determination coefficient (R²) of 0.87 and a root mean square error (RMSE) value of 97.78 for the calibration set, and an R² value of 0.91 and a RMSE value of 75.95 for the prediction set. The CARS-least-squares support vector machine model resulted in the most satisfactory predictions of Cd contents in roots, with R² values of 0.95 (RMSE, 0.92 × 10³) for the calibration set and 0.90 (RMSE, 1.64 × 10³) for the prediction set. Finally, the Cd concentrations in different plant tissues were visualized on the prediction maps by predicted spectral features on each hyperspectral image pixel. Thus, visible and near-infrared imaging combined with chemometric methods produces a promising technique to evaluate M. sacchariflorus' Cd phytoremediation capability in high-throughput metal-contaminated field applications.

Advanced Techniques for Hyperspectral Imaging in the Food Industry: Principles and Recent Applications

Hyperspectral imaging (HSI) is a technology integrating optical sensing technologies of imaging, spectroscopy, and chemometrics. The sensor of HSI can obtain both spatial and spectral information simultaneously. Therefore, the chemical and physical information of food products can be monitored in a rapid, nondestructive, and noncontact manner. There are numerous reports and papers and much research dealing with the applications of HSI in food in recent years. This review introduces the principle of HSI technology, summarizes its recent applications in food, and pinpoints future trends.

DeepSpectra: An end-to-end deep learning approach for quantitative spectral analysis

Learning patterns from spectra is critical for the development of chemometric analysis of spectroscopic data. Conventional two-stage calibration approaches consist of data preprocessing and modeling analysis. Misuse of preprocessing may introduce artifacts or remove useful patterns and result in worse model performance. An end-to-end deep learning approach incorporated Inception module, named DeepSpectra, is presented to learn patterns from raw data to improve the model performance. DeepSpectra model is compared to three CNN models on the raw data, and 16 preprocessing approaches are included to evaluate the preprocessing impact by testing four open accessed visible and near infrared spectroscopic datasets (corn, tablets, wheat, and soil). DeepSpectra model outperforms the other three convolutional neural network models on four datasets and obtains better results on raw data than in preprocessed data for most scenarios. The model is compared with linear partial least square (PLS) and nonlinear artificial neural network (ANN) methods and support vector machine (SVR) on raw and preprocessed data. The results show that DeepSpectra approach provides improved results than conventional linear and nonlinear calibration approaches in most scenarios. The increased training samples can improve the model repeatability and accuracy.

OPTICS-based Unsupervised Method for Flaking Degree Evaluation on the Murals in Mogao Grottoes

In recent years, the preventive protection and restoration work of the murals in Mogao Grottoes has received extensive attention. Due to the fragility and detachment of the murals, it is necessary to study non-contact disease detection and prevention methods. In this paper, we propose an unsupervised method to accurately predict the degree of mural flaking diseases in Mogao Grottoes. The hyperspectral image (HSI) is captured by V10-PS hyperspectral camera. The proposed method includes three main steps: (1) extract the spectral features of the HSI by Principal Component Analysis (PCA) and Sparse Auto-Encoder (SAE) respectively; (2) cluster the extracted features by the Ordering Points to Identify the Clustering Structure (OPTICS) algorithm based on the density; (3) calculate the distance between the cluster core point and the other points in the feature space and visualize the final classification result. Different from other existing hyperspectral classification works, the research proposed in this paper is the degree detection of flaking of murals. Since the degree of flaking is continuous and the work is conducted without any supervision information, the entire workflow is complex and challenging. The experimental results show the effectiveness of our method.

Rapid evaluation of the texture properties of melon (Cucumis melo L. Var. reticulata cv. Green net) using near infrared spectroscopy

The near infrared (NIR) spectroscopy as the rapid nondestructive method was aimed to be applied for determination of the texture properties of melon intact fruit and pulp including initial firmness, rupture force, average firmness, rupture distance, toughness, average penetrating force and penetrating energy. The data from the reference method of texture analyzer were correlated with the NIR spectral data. The result showed that, only the two properties including rupture force and penetrating force in pulp could be predicted by NIR spectroscopy technique. The determination coefficient of validation (r2 ) for prediction of rupture force and penetrating force in the pulp of melon using intact fruit spectra were 0.850 and 0.845, respectively. The r2 , for prediction of rupture force and penetrating force in the pulp of melon using pulp spectra were 0.813 and 0.778, respectively. This indicated that the NIR spectroscopy protocol developed here was useful for research works such as breeding and postharvest research, the melon processing factory and also the import and export of melon.

PRACTICAL APPLICATIONS

The near infrared spectroscopy protocol developed for determination of rupture force and penetrating force in pulp using intact fruit spectra as a nondestructive method will be useful for research works such as breeding and postharvest research, the melon processing factory and also the import and export of melon. There are also the protocol developed using pulp spectra can be used for texture determination of fresh-cut melon.

Application of Hyperspectral Imaging to Detect Sclerotinia sclerotiorum on Oilseed Rape Stems

Hyperspectral imaging covering the spectral range of 384-1034 nm combined with chemometric methods was used to detect Sclerotinia sclerotiorum (SS) on oilseed rape stems by two sample sets (60 healthy and 60 infected stems for each set). Second derivative spectra and PCA loadings were used to select the optimal wavelengths. Discriminant models were built and compared to detect SS on oilseed rape stems, including partial least squares-discriminant analysis, radial basis function neural network, support vector machine and extreme learning machine. The discriminant models using full spectra and optimal wavelengths showed good performance with classification accuracies of over 80% for the calibration and prediction set. Comparing all developed models, the optimal classification accuracies of the calibration and prediction set were over 90%. The similarity of selected optimal wavelengths also indicated the feasibility of using hyperspectral imaging to detect SS on oilseed rape stems. The results indicated that hyperspectral imaging could be used as a fast, non-destructive and reliable technique to detect plant diseases on stems.