A Machine Learning Method for Classification and Identification of Potato Cultivars Based on the Reaction of MOS Type Sensor-Array

Machine Learning-Based Identification of Mating Type and Metalaxyl Response in Phytophthora infestans Using SSR Markers

Phytophthora infestans is the causal agent of late blight in potato. The occurrence of P. infestans with both A1 and A2 mating types in the field may result in sexual reproduction and the generation of recombinant strains. Such strains with new combinations of traits can be highly aggressive, resistant to fungicides, and can make the disease difficult to control in the field. Metalaxyl-resistant isolates are now more prevalent in potato fields. Understanding the genetic structure and rapid identification of mating types and metalaxyl response of P. infestans in the field is a prerequisite for effective late blight disease monitoring and management. Molecular and phenotypic assays involving molecular and phenotypic markers such as mating types and metalaxyl response are typically conducted separately in the studies of the genotypic and phenotypic diversity of P. infestans. As a result, there is a pressing need to reduce the experimental workload and more efficiently assess the aggressiveness of different strains. We think that employing genetic markers to not only estimate genotypic diversity but also to identify the mating type and fungicide response using machine learning techniques can guide and speed up the decision-making process in late blight disease management, especially when the mating type and metalaxyl resistance data are not available. This technique can also be applied to determine these phenotypic traits for dead isolates. In this study, over 600 P. infestans isolates from different populations-Estonia, Pskov region, and Poland-were classified for mating types and metalaxyl response using machine learning techniques based on simple sequence repeat (SSR) markers. For both traits, random forest and the support vector machine demonstrated good accuracy of over 70%, compared to the decision tree and artificial neural network models whose accuracy was lower. There were also associations (p < 0.05) between the traits and some of the alleles detected, but machine learning prediction techniques based on multilocus SSR genotypes offered better prediction accuracy.

EfficientRMT-Net-An Efficient ResNet-50 and Vision Transformers Approach for Classifying Potato Plant Leaf Diseases

The primary objective of this study is to develop an advanced, automated system for the early detection and classification of leaf diseases in potato plants, which are among the most cultivated vegetable crops worldwide. These diseases, notably early and late blight caused by Alternaria solani and Phytophthora infestans, significantly impact the quantity and quality of global potato production. We hypothesize that the integration of Vision Transformer (ViT) and ResNet-50 architectures in a new model, named EfficientRMT-Net, can effectively and accurately identify various potato leaf diseases. This approach aims to overcome the limitations of traditional methods, which are often labor-intensive, time-consuming, and prone to inaccuracies due to the unpredictability of disease presentation. EfficientRMT-Net leverages the CNN model for distinct feature extraction and employs depth-wise convolution (DWC) to reduce computational demands. A stage block structure is also incorporated to improve scalability and sensitive area detection, enhancing transferability across different datasets. The classification tasks are performed using a global average pooling layer and a fully connected layer. The model was trained, validated, and tested on custom datasets specifically curated for potato leaf disease detection. EfficientRMT-Net's performance was compared with other deep learning and transfer learning techniques to establish its efficacy. Preliminary results show that EfficientRMT-Net achieves an accuracy of 97.65% on a general image dataset and 99.12% on a specialized Potato leaf image dataset, outperforming existing methods. The model demonstrates a high level of proficiency in correctly classifying and identifying potato leaf diseases, even in cases of distorted samples. The EfficientRMT-Net model provides an efficient and accurate solution for classifying potato plant leaf diseases, potentially enabling farmers to enhance crop yield while optimizing resource utilization. This study confirms our hypothesis, showcasing the effectiveness of combining ViT and ResNet-50 architectures in addressing complex agricultural challenges.

Determination of the effect of abiotic stress on the oxidative potential of edible potato tubers

Stress factors occurring during the growing season and potato storage, can negatively affect the quality of tubers, including an increased tendency to enzymatic darkening. Abiotic stress due to water shortage is a major factor limiting agricultural production. The purpose of the study was to determine the effect of cultivation technology based on the use of biostimulant, hydrogel and irrigation as well as storage on the propensity to darkening and the content of sugars and organic acids. The results show that genotypic and technological variability in interaction with growing season conditions had a significant (p < 0.05) effect on the oxidative potential (OP) of potato tubers. The Denar cultivar, compared to the 'Gardena', was characterized by a lower tendency to enzymatic darkening. Application of biostimulant and hydrogel generally contributed to lowering the oxidative potential of the tested cultivars. The application of anti-stress agents had no effect on organic acid content. The long-term storage caused an increase in the content of total sugars (TS) (22%), reducing sugars (RS) (49%), chlorogenic acid (ACH) (11%), and loss of ascorbic acid (AA) (6%) in the tubers which contributed to an increase in the oxidative potential of potato tubers (16%). The correlation coefficients obtained (p < 0.05) confirm the dependence of OP on the concentration of organic acids.

Environmental Engineering Applications of Electronic Nose Systems Based on MOX Gas Sensors

Nowadays, the electronic nose (e-nose) has gained a huge amount of attention due to its ability to detect and differentiate mixtures of various gases and odors using a limited number of sensors. Its applications in the environmental fields include analysis of the parameters for environmental control, process control, and confirming the efficiency of the odor-control systems. The e-nose has been developed by mimicking the olfactory system of mammals. This paper investigates e-noses and their sensors for the detection of environmental contaminants. Among different types of gas chemical sensors, metal oxide semiconductor sensors (MOXs) can be used for the detection of volatile compounds in air at ppm and sub-ppm levels. In this regard, the advantages and disadvantages of MOX sensors and the solutions to solve the problems arising upon these sensors' applications are addressed, and the research works in the field of environmental contamination monitoring are overviewed. These studies have revealed the suitability of e-noses for most of the reported applications, especially when the tools were specifically developed for that application, e.g., in the facilities of water and wastewater management systems. As a general rule, the literature review discusses the aspects related to various applications as well as the development of effective solutions. However, the main limitation in the expansion of the use of e-noses as an environmental monitoring tool is their complexity and lack of specific standards, which can be corrected through appropriate data processing methods applications.

Spotting of Volatile Signatures through GC-MS Analysis of Bacterial and Fungal Infections in Stored Potatoes (Solanum tuberosum L.)

Potatoes inoculated with Pectobacterium carotovorum spp., Aspergillus flavus and Aspergillus niger, along with healthy (control) samples, were stored at different storage temperatures (4 ± 1 °C, 8 ± 1 °C, 25 ± 1 °C) for three weeks. Volatile organic compounds (VOCs) were mapped using the headspace gas analysis through solid phase micro extraction-gas chromatography-mass spectroscopy every week. The VOC data were arranged into different groups and classified using principal component analysis (PCA) and partial least square discriminant analysis (PLS-DA) models. Based on a variable importance in projection (VIP) score > 2 and the heat map, prominent VOCs were identified as 1-butanol and 1-hexanol, which can act as biomarkers for Pectobacter related bacterial spoilage during storage of potatoes in different conditions. Meanwhile, hexadecanoic acid and acetic acid were signature VOCs for A. flavus, and hexadecane, undecane, tetracosane, octadecanoic acid, tridecene and undecene were associated with A. niger. The PLS-DA model performed better at classifying the VOCs of the three different species of infection and the control sample compared to PCA, with high values of R² (96-99%) and Q² (0.18-0.65). The model was also found to be reliable for predictability during random permutation test-based validation. This approach can be adopted for fast and accurate diagnosis of pathogenic invasion of potatoes during storage.

Determining the shelf life and quality changes of potatoes (Solanum tuberosum) during storage using electronic nose and machine learning

The activities of alpha-amylase, beta-amylase, sucrose synthase, and invertase enzymes are under the influence of storage conditions and can affect the structure of starch, as well as the sugar content of potatoes, hence altering their quality. Storage in a warehouse is one of the most common and effective methods of storage to maintain the quality of potatoes after their harvest, while preserving their freshness and sweetness. Smart monitoring and evaluation of the quality of potatoes during the storage period could be an effective approach to improve their freshness. This study is aimed at assessing the changes in the potato quality by an electronic nose (e-nose) in terms of the sugar and carbohydrate contents. Three potato cultivars (Agria, Santé, and Sprite) were analyzed and their quality variations were separately assessed. Quality parameters (i.e. sugar and carbohydrate contents) were evaluated in six 15-day periods. The e-nose data were analyzed by means of chemometric methods, including principal component analysis (PCA), linear data analysis (LDA), support vector machine (SVM), and artificial neural network (ANN). Quadratic discriminant analysis (QDA) and multivariate discrimination analysis (MDA) offer the highest accuracy and sensitivity in the classification of data. The accuracy of all methods was higher than 90%. These results could be applied to present a new approach for the assessment of the quality of stored potatoes.

Investigating Changes in pH and Soluble Solids Content of Potato during the Storage by Electronic Nose and Vis/NIR Spectroscopy

Potato is an important agricultural product, ranked as the fourth most common product in the human diet. Potato can be consumed in various forms. As customers expect safe and high-quality products, precise and rapid determination of the quality and composition of potatoes is of crucial significance. The quality of potatoes may alter during the storage period due to various phenomena. Soluble solids content (SSC) and pH are among the quality parameters experiencing alteration during the storage process. This study is thus aimed to assess the variations in SSC and pH during the storage of potatoes using an electronic nose and Vis/NIR spectroscopic techniques with the help of prediction models including partial least squares (PLS), multiple linear regression (MLR), principal component regression (PCR), support vector regression (SVR) and an artificial neural network (ANN). The variations in the SSC and pH are ascending and significant. The results also indicate that the SVR model in the electronic nose has the highest prediction accuracy for the SSC and pH (81, and 92%, respectively). The artificial neural network also managed to predict the SSC and pH at accuracies of 83 and 94%, respectively. SVR method shows the lowest accuracy in Vis/NIR spectroscopy while the PLS model exhibits the best performance in the prediction of the SSC and pH with respective precision of 89 and 93% through the median filter method. The accuracy of the ANN was 85 and 90% in the prediction of the SSC and pH, respectively.

A Green Analytical Method Combined with Chemometrics for Traceability of Tomato Sauce Based on Colloidal and Volatile Fingerprinting

Tomato sauce is a world famous food product. Despite standards regulating the production of tomato derivatives, the market suffers frpm fraud such as product adulteration, origin mislabelling and counterfeiting. Methods suitable to discriminate the geographical origin of food samples and identify counterfeits are required. Chemometric approaches offer valuable information: data on tomato sauce is usually obtained through chromatography (HPLC and GC) coupled to mass spectrometry, which requires chemical pretreatment and the use of organic solvents. In this paper, a faster, cheaper, and greener analytical procedure has been developed for the analysis of volatile organic compounds (VOCs) and the colloidal fraction via multivariate statistical analysis. Tomato sauce VOCs were analysed by GC coupled to flame ionisation (GC-FID) and to ion mobility spectrometry (GC-IMS). Instead of using HPLC, the colloidal fraction was analysed by asymmetric flow field-fractionation (AF4), which was applied to this kind of sample for the first time. The GC and AF4 data showed promising perspectives in food-quality control: the AF4 method yielded comparable or better results than GC-IMS and offered complementary information. The ability to work in saline conditions with easy pretreatment and no chemical waste is a significant advantage compared to environmentally heavy techniques. The method presented here should therefore be taken into consideration when designing chemometric approaches which encompass a large number of samples.

A Conventional VOC-PID Sensor for a Rapid Discrimination among Aromatic Plant Varieties: Classification Models Fitted to a Rosemary Case-Study

This study explores the use of a photoionization detector (PID) to distinguish varieties of rosemary plant, based on their volatile organic compound (VOC) emissions. The aim was to be able to distinguish plant varieties using a simple, quick, and inexpensive method. Two varieties were studied, Rosmarinus officinalis L. “Prostratus” and “Erectus”. First, the PID was used to detect VOCs emitted by leaves from each variety, and subsequently essential oil was extracted from the same leaves. Then, the well-established GC-MS method was used to characterize and differentiate the oil from each of the two varieties. The PID was able to capture different signals, and a ‘fingerprint’ for each of the two varieties was obtained. To validate the PID performance, the data set obtained was analyzed by means of advanced statistical models (principal component analysis, cluster and support vector machine and artificial neural network) which were able to discriminate the two varieties with high accuracy (over 80%). Therefore, the results confirm that the PID was able to detect differences in VOC emissions. In conclusion, PID proved be an interesting instrument for the classification of rosemary plants, and in this sense could be applied to other aromatic plants.

Discrimination and chemical composition quantitative model of Raw Moutan Cortex and Moutan Cortex Carbon based on electronic nose and machine learning

Raw Moutan Cortex (RMC) is a traditional medicinal material commonly used in China. Moutan Cortex Carbon (MCC) is a processed product of RMC by stir-frying. As raw and processed products of the same Chinese herb pieces, they have different effects. RMC has the effects of clearing heat and cooling blood, promoting blood circulation and removing blood stasis, but MCC has the contrary effect of cooling blood and hemostasis. Therefore, it is necessary to distinguish them effectively. The traditional quality evaluation method of RMC and MCC still adopts character identification, and mainly relies on the working experience and sensory judgment of employees with experience. This will lead to strong subjectivity and poor repeatability. And the final evaluation result may cause inevitable errors and the processed products with different processing degrees in actual production, which affects the clinical efficacy. In this study, the electronic nose technology was introduced to objectively digitize the odor of RMC and MCC. And the discrimination model of RMC and MCC was constructed in order to establish a rapid, objective and stable quality evaluation method of RMC and MCC. According to the correlation analysis, the experiment found the content of gallic acid, 5-hydroxymethylfurfural (5-HMF), paeoniflorin and paeonol determined by high performance liquid chromatography (HPLC) had a certain correlation with their odor characteristics. Thus, partial least squares regression (PLSR) and support vector machine regression (SVR) were compared and established the chemical composition quantitative model. Results showed that the quantitative data of RMC and MCC odor could be used to predict the contents of the chemical components. It can be used for quality control of RCM and MCC.

Application of MOS Gas Sensors Coupled with Chemometrics Methods to Predict the Amount of Sugar and Carbohydrates in Potatoes

Five potato varieties were studied using an electronic nose with nine MOS sensors. Parameters measured included carbohydrate content, sugar level, and the toughness of the potatoes. Routine tests were carried out while the signals for each potato were measured, simultaneously, using an electronic nose. The signals obtained indicated the concentration of various chemical components. In addition to support vector machines (SVMs that were used for the classification of the samples, chemometric methods, such as the partial least squares regression (PLSR) method, the principal component regression (PCR) method, and the multiple linear regression (MLR) method, were used to create separate regression models for sugar and carbohydrates. The predictive power of the regression models was characterized by a coefficient of determination (R²), a root-mean-square error of prediction (RMSEP), and offsets. PLSR was able to accurately model the relationship between the smells of different types of potatoes, sugar, and carbohydrates. The highest and lowest accuracy of models for predicting sugar and carbohydrates was related to Marfona potatoes and Sprite cultivar potatoes. In general, in all cultivars, the accuracy in predicting the amount of carbohydrates was somewhat better than the accuracy in predicting the amount of sugar. Moreover, the linear function had 100% accuracy for training and validation in the C-SVM method for classification of five potato groups. The electronic nose could be used as a fast and non-destructive method for detecting different potato varieties. Researchers in the food industry will find this method extremely useful in selecting the desired product and samples.

Sensors for Volatile Organic Compounds

This paper provides an overview of recent developments in the field of volatile organic compound (VOC) sensors, which are finding uses in healthcare, safety, environmental monitoring, food and agriculture, oil industry, and other fields. It starts by briefly explaining the basics of VOC sensing and reviewing the currently available and quickly progressing VOC sensing approaches. It then discusses the main trends in materials' design with special attention to nanostructuring and nanohybridization. Emerging sensing materials and strategies are highlighted and their involvement in the different types of sensing technologies is discussed, including optical, electrical, and gravimetric sensors. The review also provides detailed discussions about the main limitations of the field and offers potential solutions. The status of the field and suggestions of promising directions for future development are summarized.

An In Vitro HL-1 Cardiomyocyte-Based Olfactory Biosensor for Olfr558-Inhibited Efficiency Detection

Some short-chain fatty acids with a pungent or unpleasant odor are important components of human body odor. These malodors severely threaten human health. The antagonists of malodors would help to improve odor perception by affecting the interaction between odors and their receptors. However, the traditional odor detection and analysis methods, such as MOS, electrochemical, conductive polymer gas sensors, or chromatography-mass spectrometry are not suitable for screening the antagonists since they are unable to detect the ligand efficacy after odor-receptor binding. In this study, RT-PCR results showed that HL-1 cardiomyocytes endogenously express the olfactory receptor 558 (Olfr558) which can be activated by several malodorous short-chain fatty acids. Therefore, an in vitro HL-1 cardiomyocyte-based olfactory biosensor (HCBO-biosensor) was developed by combining cardiomyocytes and microelectrode array (MEA) chips for screening the potential antagonists of the Olfr558. Firstly, it showed that the biosensor specifically responded to ligands of Olfr558 through odor stimulation experiments. Then, an odor response model of HL-1 cardiomyocytes was constructed by a ligand of Olfr558 (isovaleric acid). The response feature of the in vitro HCBO-biosensor to individual odors and mixtures with a potential antagonist (citral or β-damascenone) were extracted and compared. Finally, the Olfr558-inhibited efficiency was indirectly detected by comparing the half-maximal inhibitory concentration of isovaleric acid. The results showed that β-damascenone greatly inhibited Olfr558 while citral showed no significant inhibitory effect. In conclusion, we built a novel screening method for the antagonists of Olfr558 based on HL-1 cardiomyocytes and the MEA chip which will assist odor-related companies to develop novel antagonists of Olfr558.

Grape Cultivar Identification and Classification by Machine Olfaction Analysis of Leaf Volatiles

Development of electronic technologies for precise identification of fruit crop cultivars in agricultural production provides an effective means for assuring product quality and authentication. The capabilities of discriminating between grape (Vitis vinifera L.) cultivars is essential for assuring certification of varieties sold in world markets. Machine olfaction, based on electronic-nose (e-nose) technologies, is readily available for rapid identification of fruit and vegetative agricultural products. This technology relies on detection of and discrimination between volatile organic compound (VOC) emissions from plant parts. It may be used in all stages of agricultural production to facilitate crop maintenance, cultivation, and harvesting decisions prior to marketing. An experimental e-nose device was constructed and tested in combination with five chemometric methods, including PCA, LDA, QDA, SVM, and ANN, as rapid, non-destructive tools for identification and classification of grape cultivars. An e-nose instrument equipped with nine metal oxide semiconductor (MOS) sensors was utilized to identify and classify five grape cultivars based on leaf VOC emissions using supervised and non-supervised methods. Grape leaf samples were first identified as belonging to specific cultivar types using PCA analyses, which are non-supervised classification methods, with the first two principal components (PC-1 and PC-2) accounting for 89% of the total variance. Four supervised statistical methods were further tested, including DA, QDA, SVM, and ANN, and provided effective discrimination accuracies of 98%, 99%, 92%, and 99%, respectively. These findings confirmed the suitable applicability of an MOS e-nose sensor array with supervised methods for accurate identification of grape cultivars, which is useful for authentication of vine cultivar types for commercial markets.

Impact of Coffee Bean Roasting on the Content of Pyridines Determined by Analysis of Volatile Organic Compounds

The aim of the study was to analyze the process of roasting coffee beans in a convection–conduction roaster (CC) without a heat exchanger and a convection–conduction–radiation roaster (CCR) with a heat exchanger for determination of the aroma profile. The aroma profile was analyzed using the SPME/GC-MS technique, and an Agrinose electronic nose was used to determine the aroma profile intensity. Arabica coffee beans from five regions of the world, namely, Peru, Costa Rica, Ethiopia, Guatemala, and Brazil, were the research material. The chemometric analyses revealed the dominance of azines, alcohols, aldehydes, hydrazides, and acids in the coffee aroma profile. Their share distinguished the aroma profiles depending on the country of origin of the coffee beans. The high content of pyridine from the azine group was characteristic for the coffee roasting process in the convection–conduction roaster without a heat exchanger, which was shown by the PCA analysis. The increased content of pyridine resulted from the appearance of coal tar, especially in the CC roaster. Pyridine has an unpleasant and bitter plant-like odor, and its excess is detrimental to the human organism. The dominant and elevated content of pyridine is a defect of the coffee roasting process in the CC roaster compared to the process carried out in the CCR machine. The results obtained with the Agrinose showed that the CC roasting method had a significant effect on the sensor responses. The effect of coal tar on the coffee beans resulted in an undesirable aroma profile characterized by increased amounts of aromatic volatile compounds and higher responses of Agrinose sensors.

Comparative flavor analysis of four kinds of sweet fermented grains by sensory analysis combined with GC-MS

Four types of cereals (glutinous rice, purple rice, red rice, yellow millet) were selected to produce sweet fermented grains. Flavor profiles of sweet fermented grains are comparatively studied to distinguish various flavor types by using GC-MS, electronic nose (E-nose), and sensory analysis, and the amino acid composition and physicochemical properties of sweet fermented grains were analyzed. The results showed that the volatile compounds of sweet fermented grains were significantly different. Esters and alcohols were the major volatile compounds in sweet fermented grains. The electronic nose, electronic tongue and sensory analysis jointly verified that the volatile components of sweet fermented grains had differences between them. The sweet fermented grains could be classified based on differences in volatile compounds. In the amino acids analysis, Glu, Pro, Asp and Leu were the most abundant. The difference in physicochemical properties is more helpful to distinguish different types of sweet fermented grains. Correlation analysis between antioxidant active substances and color value showed a positive correlation between with a* value, and a negative correlation with L*, b* value. Our results suggested that there were differences in the flavor characteristics of sweet fermented grains fermented from different types of cereals. The results of the study will provide valuable information for the selection of raw materials for sweet fermented grains.

Comparison of Cheese Aroma Intensity Measured Using an Electronic Nose (E-Nose) Non-Destructively with the Aroma Intensity Scores of a Sensory Evaluation: A Pilot Study

Cheese aroma is known to affect consumer preference. One of the methods to measure cheese aroma is the use of an electronic nose (e-nose), which has been used to classify cheese types, production areas, and cheese ages. However, few studies have directly compared the aroma intensity scores derived from sensory evaluations with the values of metal oxide semiconductor sensors that can easily measure the aroma intensity. This pilot study aimed to investigate the relationship between sensory evaluation scores and e-nose values with respect to cheese aroma. Five types of processed cheese (two types of normal processed cheese, one type containing aged cheese, and two types containing blue cheese), and one type of natural cheese were used as samples. The sensor values obtained using the electronic nose, which measured sample aroma non-destructively, and five sensory evaluation scores related to aroma (aroma intensity before intake, during mastication, and after swallowing; taste intensity during mastication; and remaining flavor after swallowing (lasting flavor)) determined by six panelists, were compared. The e-nose values of many of the tested cheese types were significantly different, whereas the sensory scores of the one or two types of processed cheese containing blue cheese and those of the natural cheese were significantly different. Significant correlations were observed between the means of e-nose values and the medians of aroma intensity scores derived from the sensory evaluation testing before intake, during mastication, and after swallowing. In particular, the aroma intensity score during mastication was found to have a linear relationship with the e-nose values (Pearson’s R = 0.983). In conclusion, the e-nose values correlated with the sensory scores with respect to cheese aroma intensity and could be helpful in predicting them.

Recent Progress in Smart Electronic Nose Technologies Enabled with Machine Learning Methods

Machine learning methods enable the electronic nose (E-Nose) for precise odor identification with both qualitative and quantitative analysis. Advanced machine learning methods are crucial for the E-Nose to gain high performance and strengthen its capability in many applications, including robotics, food engineering, environment monitoring, and medical diagnosis. Recently, many machine learning techniques have been studied, developed, and integrated into feature extraction, modeling, and gas sensor drift compensation. The purpose of feature extraction is to keep robust pattern information in raw signals while removing redundancy and noise. With the extracted feature, a proper modeling method can effectively use the information for prediction. In addition, drift compensation is adopted to relieve the model accuracy degradation due to the gas sensor drifting. These recent advances have significantly promoted the prediction accuracy and stability of the E-Nose. This review is engaged to provide a summary of recent progress in advanced machine learning methods in E-Nose technologies and give an insight into new research directions in feature extraction, modeling, and sensor drift compensation.

Characteristic-Aroma-Component-Based Evaluation and Classification of Strawberry Varieties by Aroma Type

The unique fruity aroma of strawberries, a popular fruit of high economic value, is closely related to all the volatile organic compounds (VOCs) contained within them. Despite extensive studies on the identification of VOCs in strawberries, systematic studies on fruit-aroma-related VOCs are few, resulting in a lack of effective standards for accurately distinguishing aroma types. In the present study, solid-phase micro extraction and gas chromatography–mass spectrometry were used to analyze and identify VOCs in the ripe fruit of each of the 16 strawberry varieties at home and abroad and to explore their characteristic aroma components and the classification of such varieties by aroma type. The results suggested remarkable variations in the types and contents of VOCs in different strawberry varieties, of which esters were dominant. The principal volatile components, consisting of four esters, three alcohols, one aldehyde, and one ketone, in 16 strawberry varieties were detected based on the absolute and relative contents of VOCs in the fruit. The characteristic aroma components in strawberries, containing nine esters, six aldehydes, and one alcohol, were determined based on the aroma values of different VOCs, and the characteristic aroma components were divided into five types further based on aroma descriptions. Sixteen strawberry varieties were finally divided into four aroma types, namely, peachy, pineapple, fruity, and floral, based on the contributions of different types. The results provided a basis and standard for classifying strawberries by aroma type, studying the hereditary regularity of the fruity aroma of strawberries, and improving aroma quality.