Standardisation of near infrared hyperspectral imaging for quantification and classification of DON contaminated wheat samples

Predicting the oil content of individual corn kernels combining NIR-HSI and multi-stage parameter optimization techniques

Predicting the oil content of individual corn kernels using hyperspectral imaging and ML offers the advantages of being rapid and non-destructive. However, traditional methods rely on expert experience for setting parameters. In response to these limitations, this study has designed an innovative multi-stage grid search technique, tailored to the characteristics of spectral data. Initially, the study automatically screening the best model from up to 504 algorithm combinations. Subsequently, multi-stage grid search is utilized for improving precision. We collected 270 kernel samples from different parts of the ear from 15 high oil and regular corn materials, with oil contents ranging from 1.4% to 13.1%. Experimental results show that the combinations SG + NONE+KS + PLSR(R2: 0.8570) and MA + LAR+Random+MLR(R2: 0.8523) performed optimally. After parameter optimization, their R2 values increased to 0.9045 and 0.8730, respectively. Additionally, the ACNNR model achieved an R2 of 0.8878 and an RMSE of 0.2243. The improved algorithm significantly outperforms traditional methods and ACNNR model in prediction accuracy and adaptability, offering an effective method for field applications.

NIR-HSI as a tool to predict deoxynivalenol and fumonisins in maize kernels: a step forward in preventing mycotoxin contamination

BACKGROUND

Maize is frequently contaminated with deoxynivalenol (DON) and fumonisins B1 (FB1) and B2 (FB2). In the European Union, these mycotoxins are regulated in maize and maize-derived products. To comply with these regulations, industries require a fast, economic, safe, non-destructive and environmentally friendly analysis method.

RESULTS

In the present study, near-infrared hyperspectral imaging (NIR-HSI) was used to develop regression and classification models for DON, FB1 and FB2 in maize kernels. The best regression models presented the following root mean square error of cross validation and ratio of performance to deviation values: 0.848 mg kg-1 and 2.344 (DON), 3.714 mg kg-1 and 2.018 (FB1) and 2.104 mg kg-1 and 2.301 (FB2). Regarding classification, European Union legal limits for DON and FB1 + FB2 were selected as thresholds to classify maize kernels as acceptable or not. The sensitivity and specificity were 0.778 and 1 for the best DON classification model and 0.607 and 0.938 for the best FB1 + FB2 classification model.

CONCLUSION

NIR-HSI can help reduce DON and fumonisins contamination in the maize food and feed chain. © 2024 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Advancements, limitations and challenges in hyperspectral imaging for comprehensive assessment of wheat quality: An up-to-date review

The potential of hyperspectral imaging technology (HIT) for the determination of physicochemical and nutritional components, evaluation of fungal/mycotoxins contamination, wheat varieties classification, identification of non-mildew-damaged wheat kernels, as well as detection of flour adulteration is comprehensively illustrated and reviewed. The latest findings (2018-2023) of HIT in wheat quality evaluation through internal and external attributes are compared and summarized in detail. The limitations and challenges of HIT to improve assessment accuracy are clearly described. Additionally, various practical recommendations and strategies for the potential application of HIT are highlighted. The future trends and prospects of HIT in evaluating wheat quality are also mentioned. In conclusion, HIT stands as a cutting-edge technology with immense potential for revolutionizing wheat quality evaluation. As advancements in HIT continue, it will play a pivotal role in shaping the future of wheat quality assessment and contributing to a more sustainable and efficient food supply chain.

Near infrared hyperspectral imaging as a sorting tool for deoxynivalenol reduction in wheat batches

The present study aimed to evaluate the feasibility of using near-infrared hyperspectral imaging (NIR-HSI) and chemometrics for classification of individual wheat kernels according to their deoxynivalenol (DON) level. In total, 600 wheat kernels from samples naturally contaminated over the maximum EU level were collected, and the DON content in each individual wheat kernel was analyzed by UHPLC. Linear discriminant analysis (LDA) was employed for building classification models of DON using the EU maximum level as cut off level, and they were tested on balanced and imbalanced test sets. The results showed that the models presented a balanced accuracy of 0.71, that would allow to obtain safe batches from contaminated batches once the unsafe kernels had been rejected, but often more than 30% of the batch would be rejected. The work confirmed that NIR-HSI could be a feasible method for monitoring DON in individual kernels and removing highly contaminated kernels prior to food chain entry.

Enhancing traceability of wheat quality through the supply chain

With the growing global population, the need for food is expected to grow tremendously in the next few decades. One of the key tools to address such growing food demand is minimizing grain losses and optimizing food processing operations. Hence, several research studies are underway to reduce grain losses/degradation at the farm (upon harvest) and later during the milling and baking processes. However, less attention has been paid to changes in grain quality between harvest and milling. This paper aims to address this knowledge gap and discusses possible strategies for preserving grain quality (for Canadian wheat in particular) during unit operations at primary, process, or terminal elevators. To this end, the importance of wheat flour quality metrics is briefly described, followed by a discussion on the effect of grain properties on such quality parameters. This work also explores how drying, storage, blending, and cleaning, as some of the common post-harvest unit operations, could affect grain's end-product quality. Finally, an overview of the available techniques for grain quality monitoring is provided, followed by a discussion on existing gaps and potential solutions for quality traceability throughout the wheat supply chain.

A Review of Recent Advances for the Detection of Biological, Chemical, and Physical Hazards in Foodstuffs Using Spectral Imaging Techniques

Traditional methods for detecting foodstuff hazards are time-consuming, inefficient, and destructive. Spectral imaging techniques have been proven to overcome these disadvantages in detecting foodstuff hazards. Compared with traditional methods, spectral imaging could also increase the throughput and frequency of detection. This study reviewed the techniques used to detect biological, chemical, and physical hazards in foodstuffs including ultraviolet, visible and near-infrared (UV-Vis-NIR) spectroscopy, terahertz (THz) spectroscopy, hyperspectral imaging, and Raman spectroscopy. The advantages and disadvantages of these techniques were discussed and compared. The latest studies regarding machine learning algorithms for detecting foodstuff hazards were also summarized. It can be found that spectral imaging techniques are useful in the detection of foodstuff hazards. Thus, this review provides updated information regarding the spectral imaging techniques that can be used by food industries and as a foundation for further studies.

Quantitative assessment of adulteration of coconut oil using transmittance multispectral imaging

Economical to a fault, coconut oil is a commodity related to fraudulent activities such as oil adulteration for undue profits. Unfortunately, the conventional methods used in the detection of adulteration and toxicants are laborious, destructive, and time-consuming. Hence, it is imperative to engineer a non-destructive and rapid screening test with sufficient accuracy. To that end, the proposed work has an in-house developed imaging system hardware and a method to estimate relevant quality parameters from multispectral imagery. Multispectral images of adulterated coconut oil were analyzed through a cascade of statistical algorithms: Fisher Discriminant Analysis and Bhattacharyya distance respectively. In this work, a functional relationship was developed for the estimation of adulteration level that recorded an R2 of 0.9876 for the training samples and an MSE of 0.0029 for the testing samples. Besides, the proposed imaging system offers flexibility on post-processing of raw measurements as the algorithm is designed to operate from raw multispectral images. In addition, the developed imaging system is economical in its capacity to estimate the adulteration of coconut oil with remarkable accuracy considering the low cost of production. Moreover, the proposed work validates the use of multispectral imagery as an initial screening technique instead of expensive spectroscopy methods.

Quantification and classification of deoxynivalenol-contaminated oat samples by near-infrared hyperspectral imaging

Deoxynivalenol (DON) is the most occurring mycotoxin in oat and oat-based products. Near-infrared hyperspectral imaging (NIR-HSI) has been proposed as a promising methodology for analysing DON contamination in the food industry. The present study aims to apply NIR-HSI for DON detection in oat kernels and to quantify and classify naturally DON-contaminated oat samples. Unground and ground oat samples were scanned by NIR-HSI before their DON content was determined by HPLC. The data were pre-treated and analysed by PLS regression and four classification methods. The most efficient DON prediction model was for unground samples (R² = 0.75 and RMSEP = 403.18 μg/kg), using twelve characteristic wavelengths with a special interest in 1203 and 1388 nm. The random forest algorithm of unground samples according to the EU maximum limit for unprocessed oats (1750 μg/kg) achieved a classification accuracy of 77.8 %. These findings indicate that NIR-HSI is a promising tool for detecting DON in oats.

Detection of wheat Fusarium head blight using UAV-based spectral and image feature fusion

Infection caused by Fusarium head blight (FHB) has severely damaged the quality and yield of wheat in China and threatened the health of humans and livestock. Inaccurate disease detection increases the use cost of pesticide and pollutes farmland, highlighting the need for FHB detection in wheat fields. The combination of spectral and spatial information provided by image analysis facilitates the detection of infection-related damage in crops. In this study, an effective detection method for wheat FHB based on unmanned aerial vehicle (UAV) hyperspectral images was explored by fusing spectral features and image features. Spectral features mainly refer to band features, and image features mainly include texture and color features. Our aim was to explain all aspects of wheat infection through multi-class feature fusion and to find the best FHB detection method for field wheat combining current advanced algorithms. We first evaluated the quality of the two acquired UAV images and eliminated the excessively noisy bands in the images. Then, the spectral features, texture features, and color features in the images were extracted. The random forest (RF) algorithm was used to optimize features, and the importance value of the features determined whether the features were retained. Feature combinations included spectral features, spectral and texture features fusion, and the fusion of spectral, texture, and color features to combine support vector machine, RF, and back propagation neural network in constructing wheat FHB detection models. The results showed that the model based on the fusion of spectral, texture, and color features using the RF algorithm achieved the best performance, with a prediction accuracy of 85%. The method proposed in this study may provide an effective way of FHB detection in field wheat.

Effects of Macroautophagy and Mitophagy on the Pathogenicity of Fusarium graminearum

Fusarium graminearum is the main pathogen of Fusarium head blight (FHB), which causes huge economic losses every year. In this study, an attempt was made to control FHB from the point of view of the physiological behavior of the pathogen itself. Autophagic inhibitors and activators were used, and the pathogenicity-related indices of F. graminearum were measured. The results showed that under nitrogen-rich conditions, macroautophagy inhibition and activation greatly reduced the mycelium weight to 0.28 and 0.25 g/ml at 24 h, which were 17.82 and 24.77% lower than that of the control treatment, respectively. Mitophagy inhibition also significantly decreased the mycelium weight (P < 0.05). Conidial yield was found to be affected by factors related to autophagy occurrence. It was found that both autophagy inhibition and activation could reduce the conidiation of F. graminearum. The toxin contents in wheat medium of macroautophagy activation treatments were 0.678, 0.190, 0.402, and 0.195 μg/g when cultured for 8 and 24 h under 0% N and 100% N conditions, respectively, which were significantly higher than those of the control treatments (P < 0.05). The infection length was measured to characterize the infectivity of F. graminearum, and we found that the length was short under macroautophagy activation conditions. However, mitophagy did not seem to affect the infectivity of F. graminearum. In summary, the above results indicate that macroautophagy and mitophagy inhibition could reduce the pathogenicity of F. graminearum, which may provide a new perspective for management of plant fungal diseases.

As the number falls, alternatives to the Hagberg-Perten falling number method: A review

This review examines the application, limitations, and potential alternatives to the Hagberg-Perten falling number (FN) method used in the global wheat industry for detecting the risk of poor end-product quality mainly due to starch degradation by the enzyme α-amylase. By viscometry, the FN test indirectly detects the presence of α-amylase, the primary enzyme that digests starch. Elevated α-amylase results in low FN and damages wheat product quality resulting in cakes that fall, and sticky bread and noodles. Low FN can occur from preharvest sprouting (PHS) and late maturity α-amylase (LMA). Moist or rainy conditions before harvest cause PHS on the mother plant. Continuously cool or fluctuating temperatures during the grain filling stage cause LMA. Due to the expression of additional hydrolytic enzymes, PHS has a stronger negative impact than LMA. Wheat grain with low FN/high α-amylase results in serious losses for farmers, traders, millers, and bakers worldwide. Although blending of low FN grain with sound wheat may be used as a means of moving affected grain through the marketplace, care must be taken to avoid grain lots from falling below contract-specified FN. A large amount of sound wheat can be ruined if mixed with a small amount of sprouted wheat. The FN method is widely employed to detect α-amylase after harvest. However, it has several limitations, including sampling variability, high cost, labor intensiveness, the destructive nature of the test, and an inability to differentiate between LMA and PHS. Faster, cheaper, and more accurate alternatives could improve breeding for resistance to PHS and LMA and could preserve the value of wheat grain by avoiding inadvertent mixing of high- and low-FN grain by enabling testing at more stages of the value stream including at harvest, delivery, transport, storage, and milling. Alternatives to the FN method explored here include the Rapid Visco Analyzer, enzyme assays, immunoassays, near-infrared spectroscopy, and hyperspectral imaging.

Rapid Foreign Object Detection System on Seaweed Using VNIR Hyperspectral Imaging

The consumption of seaweed is increasing year by year worldwide. Therefore, the foreign object inspection of seaweed is becoming increasingly important. Seaweed is mixed with various materials such as laver and sargassum fusiforme. So it has various colors even in the same seaweed. In addition, the surface is uneven and greasy, causing diffuse reflections frequently. For these reasons, it is difficult to detect foreign objects in seaweed, so the accuracy of conventional foreign object detectors used in real manufacturing sites is less than 80%. Supporting real-time inspection should also be considered when inspecting foreign objects. Since seaweed requires mass production, rapid inspection is essential. However, hyperspectral imaging techniques are generally not suitable for high-speed inspection. In this study, we overcome this limitation by using dimensionality reduction and using simplified operations. For accuracy improvement, the proposed algorithm is carried out in 2 stages. Firstly, the subtraction method is used to clearly distinguish seaweed and conveyor belts, and also detect some relatively easy to detect foreign objects. Secondly, a standardization inspection is performed based on the result of the subtraction method. During this process, the proposed scheme adopts simplified and burdenless calculations such as subtraction, division, and one-by-one matching, which achieves both accuracy and low latency performance. In the experiment to evaluate the performance, 60 normal seaweeds and 60 seaweeds containing foreign objects were used, and the accuracy of the proposed algorithm is 95%. Finally, by implementing the proposed algorithm as a foreign object detection platform, it was confirmed that real-time operation in rapid inspection was possible, and the possibility of deployment in real manufacturing sites was confirmed.

Research advancements in optical imaging and spectroscopic techniques for nondestructive detection of mold infection and mycotoxins in cereal grains and nuts

Cereal grains and nuts are represented as the economic backbone of many developed and developing countries. Kernels of cereal grains and nuts are prone to mold infection under high relative humidity and suitable temperature conditions in the field as well as storage conditions. Health risks caused by molds and their molecular metabolite mycotoxins are, therefore, important topics to investigate. Strict regulations have been developed by international trade regulatory bodies for the detection of mold growth and mycotoxin contamination across the food chain starting from the harvest to storage and consumption. Molds and aflatoxins are not evenly distributed over the bulk of grains, thus appropriate sampling for detection and quantification is crucial. Existing reference methods for mold and mycotoxin detection are destructive in nature as well as involve skilled labor and hazardous chemicals. Also, these methods cannot be used for inline sorting of the infected kernels. Thus, analytical methods have been extensively researched to develop the one that is more practical to be used in commercial detection and sorting processes. Among various analytical techniques, optical imaging and spectroscopic techniques are attracting growers' attention for their potential of nondestructive and rapid inline identification and quantification of molds and mycotoxins in various food products. This review summarizes the recent application of rapid and nondestructive optical imaging and spectroscopic techniques, including digital color imaging, X-ray imaging, near-infrared spectroscopy, fluorescent, multispectral, and hyperspectral imaging. Advance chemometric techniques to identify very low-level mold growth and mycotoxin contamination are also discussed. Benefits, limitations, and challenges of deploying these techniques in practice are also presented in this paper.

Evaluation of Dianhong black tea quality using near-infrared hyperspectral imaging technology

BACKGROUND

Tea (Camellia sinensis L) is a highly nutritious beverage with commercial value globally. However, it is at risk of economic fraud. This study aims to develop a powerful evaluation method to distinguish Chinese official Dianhong tea from various other categories, employing hyperspectral imaging (HSI) technology and chemometric algorithms.

RESULTS

Two matrix statistical algorithms encompassing a gray-level co-occurrence matrix (GLCM) and a gradient co-occurrence matrix (GLGCM) are used to extract HSI texture data. Three novel spectral variable screening methods are utilized to select wavenumbers of near-infrared (NIR) spectra: iteratively retaining informative variables (IRIV), interval random frog, and variable combination population analysis. Feature fusion of image texture characteristics and spectra data are the eigenvectors for model building. Authentic classification models are constructed using the extreme learning machine approach and the least squares support vector machine (LSSVM) approach, coupling them with features from wavelength extraction techniques for assessing the quality of Dianhong black tea. The results demonstrate that the LSSVM model using fused data (IRIV + GLGCM) provides the best results and achieves a predictive precision of 99.57%.

CONCLUSION

This study confirms that HSI coupled with LSSVM is effective in differentiating authentic Dianhong black tea samples. © 2020 Society of Chemical Industry.

Assessment of the vigor of rice seeds by near-infrared hyperspectral imaging combined with transfer learning

Rice seed vigor plays a significant role in determining the quality and quantity of rice production. Thus, the quick and non-destructive identification of seed vigor is not only beneficial to fully obtain the state of rice seeds but also the intelligent development of agriculture by instant monitoring. Thus, herein, near-infrared hyperspectral imaging technology, as an information acquisition tool, was introduced combined with a deep learning algorithm to identify the rice seed vigor. Both the spectral images and average spectra of the rice seeds were sent to discriminant models including deep learning models and traditional machine learning models, and the highest accuracy of vigor identification reached 99.5018% using the self-built model. The parameters of the established deep learning models were frozen to be feature extractor for transfer learning. The identification results whose highest number also reached almost 98% indicated the possibility of applying transfer learning to improve the universality of the models. Moreover, by visualizing the output of convolutional layers, the progress and mechanism of spectral image feature extraction in the established deep learning model was explored. Overall, the self-built deep learning models combined with near-infrared hyperspectral images in the determination of rice seed vigor have potential to efficiently perform this task.

Standardization of near infrared hyperspectral imaging for wheat single kernel sorting according to deoxynivalenol level

The spatial recognition feature of near infrared hyperspectral imaging (HSI-NIR) makes it potentially suitable for Fusarium and deoxynivalenol (DON) management in single kernels to break with heterogeneity of contamination in wheat batches to move towards individual kernel sorting and provide more quick, environmental-friendly and non-destructive analysis than wet-chemistry techniques. The aim of this study was to standardize HSI-NIR for individual kernel analysis of Fusarium damage and DON presence, to predict the level of contamination and classify grains according to the EU maximum limit (1250 µg/kg). Visual inspection on Fusarium infection symptoms and HPLC analysis for DON determination were used as reference methods. The kernels were scanned in both crease-up and crease-down position and for different image captures. The spectra were pretreated by Multiplicative Scatter Correction (MSC) and Standard Normal Variate (SNV), 1st and 2nd derivatives and normalisation, and they were evaluated also by removing spectral tails. The best fitted predictive model was on SNV pretreated data (R² 0.88 and RMSECV 4.8 mg/kg) in which 7 characteristic wavelengths were used. Linear Discriminant Analysis (LDA), Naïve Bayes and K-nearest Neighbours models classified with 100% of accuracy 1st derivative and SNV pretreated spectra according to symptomatology and with 98.9 and 98.4% of correctness 1st derivative and SNV spectra, respectively. The starting point results are encouraging for future investigations on HSI-NIR technique application to Fusarium and DON management in single wheat kernels to overcome their contamination heterogeneity.

Hyperspectral imaging and improved feature variable selection for automated determination of deoxynivalenol in various genetic lines of barley kernels for resistance screening

Fusarium head blight (FHB), a fungus disease of small grain cereal crops, results in reduced yields and diminished value of harvested grain due to the presence of deoxynivalenol (DON), a mycotoxin produced by the causal pathogen Fusarium graminearum. DON and other tricothecene mycotoxins pose serious health risks to both humans and livestock, especially swine. Due to these health concerns, barley used for malting, food or feed is routinely assayed for DON levels. Various methods are available for assaying DON levels in grain samples including enzyme-linked immunosorbent assay (ELISA) and gas chromatography-mass spectrometry (GC-MS). ELISA and GC-MS are very accurate; however, assaying grain samples by these techniques are laborious, expensive and destructive. In this study, we explored the feasibility of using hyperspectral imaging (382-1030 nm) to develop a rapid and non-destructive protocol for assaying DON in barley kernels. Samples of 888 and 116 from various genetic lines were selected for calibration and prediction. Full-wavelength locally weighted partial least squares regression (LWPLSR) achieved high accuracy with the coefficient of determination in prediction (R²_P) of 0.728 and root mean square error of prediction (RMSEP) of 3.802. Competitive adaptive reweighted sampling (CARS) was used to choose potential feature wavelengths, and these selected variables were further optimized using the iterative selection of successive projections algorithm (ISSPA). The CARS-ISSPA-LWPLSR model developed using 7 feature variables yielded R²_P of 0.680 and RMSEP of 4.213 in DON content prediction. Based on the 7 wavelengths selected by CARS-ISSPA, partial least square discriminant analysis (PLSDA) discriminated barley kernels having lower DON (less than1.25 mg/kg) levels from those with higher levels (including 1.25-3 mg/kg, 3-5 mg/kg, and 5-10 mg/kg), with Matthews correlation coefficient in cross-validation (M-R_CV) of as high as 0.931. The results demonstrate that hyperspectral imaging have potential for accelerating non-destructive DON assays of barley samples.

Near-infrared hyperspectral imaging for deoxynivalenol and ergosterol estimation in wheat samples

The present study aimed to evaluate the use of hyperspectral imaging (HSI)-NIR spectroscopy to assess the presence of DON and ergosterol in wheat samples through prediction and classification models. To achieve these objectives, a first set of bulk samples was scanned by HSI-NIR and divided into two subsamples, one that was analysed for ergosterol and another that was analysed for DON by HPLC. This method was repeated for a second larger set to build prediction and classification models. All the spectra were pretreated and statistically processed by PLS and LDA. The prediction models presented a RMSEP of 1.17 mg/kg and 501 µg/kg for ergosterol and DON, respectively. Classification achieved an encouraging accuracy of 85.4% for an independent validation set of samples. The results confirm that HSI-NIR may be a suitable technique for ergosterol quantification and DON classification of samples according to the EU legal limit for DON.

Identification of Bacterial Blight Resistant Rice Seeds Using Terahertz Imaging and Hyperspectral Imaging Combined With Convolutional Neural Network

Because bacterial blight (BB) disease seriously affects the yield and quality of rice, breeding BB resistant rice is an important priority for plant breeders but the process is time-consuming. The feasibility of using terahertz imaging technology and near-infrared hyperspectral imaging technology to identify BB resistant seeds has therefore been studied. The two-dimensional (2D) spectral images and one-dimensional (1D) spectra provided by both imaging methods were used to build discriminant models based on a deep learning method, the convolutional neural network (CNN), and traditional machine learning methods, support vector machine (SVM), random forest (RF), and partial least squares discriminant analysis (PLS-DA). The highest classification accuracy was achieved by the discriminate model based on CNN using the terahertz absorption spectra. Confusion matrixes were pictured to show the identification details. The t-distributed stochastic neighbor embedding (t-SNE) method was used to visualize the process of CNN data processing. Terahertz imaging technology combined with CNN has great potential to quickly identify BB resistant rice seeds and is more accurate than using near-infrared hyperspectral imaging.