Online Inspection of Browning in Yali Pears Using Visible-Near Infrared Spectroscopy and Interpretable Spectrogram-Based CNN Modeling

Online inspection of blackheart in potatoes using visible-near infrared spectroscopy and interpretable spectrogram-based modified ResNet modeling

Introduction

Blackheart is one of the most common physiological diseases in potatoes during storage. In the initial stage, black spots only occur in tissues near the potato core and cannot be detected from an outward appearance. If not identified and removed in time, the disease will seriously undermine the quality and sale of theentire batch of potatoes. There is an urgent need to develop a method for early detection of blackheart in potatoes.

Methods

This paper used visible-near infrared (Vis/NIR) spectroscopy to conduct online discriminant analysis on potatoes with varying degrees of blackheart and healthy potatoes to achieve real-time detection. An efficient and lightweight detection model was developed for detecting different degrees of blackheart in potatoes by introducing the depthwise convolution, pointwise convolution, and efficient channel attention modules into the ResNet model. Two discriminative models, the support vector machine (SVM) and the ResNet model were compared with the modified ResNet model.

Results and discussion

The prediction accuracy for blackheart and healthy potatoes test sets reached 0.971 using the original spectrum combined with a modified ResNet model. Moreover, the modified ResNet model significantly reduced the number of parameters to 1434052, achieving a substantial 62.71% reduction in model complexity. Meanwhile, its performance was evidenced by a 4.18% improvement in accuracy. The Grad-CAM++ visualizations provided a qualitative assessment of the model's focus across different severity grades of blackheart condition, highlighting the importance of different wavelengths in the analysis. In these visualizations, the most significant features were predominantly found in the 650-750 nm range, with a notable peak near 700 nm. This peak was speculated to be associated with the vibrational activities of the C-H bond, specifically the fourth overtone of the C-H functional group, within the molecular structure of the potato components. This research demonstrated that the modified ResNet model combined with Vis/NIR could assist in the detection of different degrees of black in potatoes.

Rapid Cooling Delays the Occurring of Core Browning in Postharvest 'Yali' Pear at Advanced Maturity by Inhibiting Ethylene Metabolism

During the storage and transportation processes, the occurrence of core browning in 'Yali' pear fruit due to adversity injury can be easily mitigated by implementing different cooling methods, especially in advanced maturity fruits. In this study, 'Yali' pears at an advanced maturity stage were subjected to slow cooling and rapid cooling treatment. The quality-related physiological percentage and severity, and the rate of good fruits were determined, and RNA-seq was used to explore the effects of different cooling methods on pathways related to core browning in advanced-maturity pears at the transcriptional level. The results indicated that, compared with slow cooling treatment, rapid cooling significantly inhibited core browning in advanced-maturity 'Yali' pears. Measurements of quality-related physiological indexes suggested that rapid cooling treatment led to higher SSC content, firmness, L* value, and b* value, indicating better brightness, coloration, and higher soluble solid content, which are desirable for commercial sale. Rapid cooling effectively suppressed the physiological metabolism of 'Yali' pears, delaying fruit senescence compared with slow-cooling treatment. Furthermore, the RNA-Seq sequencing results revealed that pathways related to browning are involved in hormone signal transduction pathways, which are associated with resistance and aging processes of pear fruit. In summary, rapid cooling treatment delayed the core browning of advanced maturity of 'Yali' pears, indicating that the core browning of 'Yali' pears is related to the cooling method, and the mechanism of rapid cooling in reducing the core browning of advanced maturity of 'Yali' pears was by delaying the aging process of the fruit. This provides a new perspective for alleviating the core browning of advanced-maturity 'Yali' pears during storage and transportation, and provides a theoretical reference for studying the mechanism of core browning of 'Yali' pears.

A low-cost discrete Vis-NIR optical sensing method for the determination of pear internal blackheart

In this study, a moldy crown pear core detection system based on a micro-optical sensor was developed. The micro-optical sensor has seven specific wavelengths, 425, 455, 515, 615, 660, 700, and 850 nm, and a cost-effective advantage. For the discrete spectrum, 7 kinds of preprocessing methods were compared. Traditional preprocessing methods, such as the standard normal transform (SNV) and multiple scattering correction (MSC) methods, cannot improve the efficiency of the spectrum. It was verified that the Savitzky - Golay (SG) convolution smoothing preprocessing method could be applied to preprocess discrete spectral data. The correlation of the spectrum after SG preprocessing in the partial least squares regression (PLSR) prediction model was 0.86, and the root mean square error (RMSE) was 0.19. Furthermore, the difference between the nonlinear modeling method without preprocessing and the PLS prediction model after preprocessing was compared. The results showed that the accuracy of the nonlinear modeling method for the discrete spectrum was much higher than that of the PLS linear modeling. The average model accuracy was above 0.9, and the k nearest neighbor (KNN) algorithm had the best effect, reaching an accuracy of 0.96. Finally, a prediction model accuracy of 0.98 was obtained by combining SG + KNN. In summary, the micro-optical sensor system had the advantages of low-cost performance and high precision, which are convenient for popularization and application in practice.

Improving prediction model robustness with virtual sample construction for near-infrared spectra analysis

In a qualitative analysis of near-infrared spectroscopy (NIRS), when the samples to be analyzed are difficult to obtain or there are few counterexamples, the robustness of the models is poor, resulting in the decline of the generalization ability of the models. In this case, the effective method is to construct virtual samples to achieve the balance of categories. In this contribution, three virtual spectrum construction strategies including Synthetic Minority Oversampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN), and Deep Convolutional Generative Adversarial Network (DCGAN) were explored to deal with the problem of insufficient or imbalanced sample numbers in NIRS analysis. The strategies were tested with the melamine and Yali pears two spectral datasets. The PLS-DA and Correct Recognition Rate (CRR) were used for discriminant model construction and accuracy evaluation, respectively. The results show that SMOTE, ADASYN, and DCGAN processing strategies can all improve the global CRR (CRRglob). The SMOTE and ADASYN can improve the CRR for majority class sample (CRRmaj), but the CRR for minority class sample (CRRmin) has decreased. For the DCGAN method, the CRRglob, CRRmaj, and CRRmin were all improved. The standard deviation of the results of the multiple parallel calculations demonstrates the robustness of DCGAN generation method. Therefore, the DCGAN method has good reliability and practicability, and can increase the robustness and generalization ability of the NIRS model.

Classification of Food Additives Using UV Spectroscopy and One-Dimensional Convolutional Neural Network

Food additives are utilized in countless food products available for sale. They enhance or obtain a specific flavor, extend the storage time, or obtain a desired texture. This paper presents an automatic classification system for five food additives based on their absorbance in the ultraviolet domain. Solutions with different concentrations were created by dissolving a measured additive mass into distilled water. The analyzed samples were either simple (one additive solution) or mixed (two additive solutions). The substances presented absorbance peaks between 190 nm and 360 nm. Each substance presents a certain number of absorbance peaks at specific wavelengths (e.g., acesulfame potassium presents an absorbance peak at 226 nm, whereas the peak associated with potassium sorbate is at 254 nm). Therefore, each additive has a distinctive spectrum that can be used for classification. The sample classification was performed using deep learning techniques. The samples were associated with numerical labels and divided into three datasets (training, validation, and testing). The best classification results were obtained using CNN (convolutional neural network) models. The classification of the 404 spectra with a CNN model with three convolutional layers obtained a mean testing accuracy of 92.38% ± 1.48%, whereas the mean validation accuracy was 93.43% ± 2.01%.

Rapid Nondestructive Testing Technology-Based Biosensors for Food Analysis

Food analysis plays a vital role in ensuring the safety and quality of food products [...].