Classification and Discrimination of Different Fungal Diseases of Three Infection Levels on Peaches Using Hyperspectral Reflectance Imaging Analysis

A dual-branch selective attention capsule network for classifying kiwifruit soft rot with hyperspectral images

Kiwifruit soft rot is highly contagious and causes serious economic loss. Therefore, early detection and elimination of soft rot are important for postharvest treatment and storage of kiwifruit. This study aims to accurately detect kiwifruit soft rot based on hyperspectral images by using a deep learning approach for image classification. A dual-branch selective attention capsule network (DBSACaps) was proposed to improve the classification accuracy. The network uses two branches to separately extract the spectral and spatial features so as to reduce their mutual interference, followed by fusion of the two features through the attention mechanism. Capsule network was used instead of convolutional neural networks to extract the features and complete the classification. Compared with existing methods, the proposed method exhibited the best classification performance on the kiwifruit soft rot dataset, with an overall accuracy of 97.08% and a 97.83% accuracy for soft rot. Our results confirm that potential soft rot of kiwifruit can be detected using hyperspectral images, which may contribute to the construction of smart agriculture.

Portable and low-cost hologram verification module using a snapshot-based hyperspectral imaging algorithm

One of the challenges in differentiating a duplicate hologram from an original one is reflectivity. A slight change in lighting condition will completely change the reflection pattern exhibited by a hologram, and consequently, a standardized duplicate hologram detector has not yet been created. In this study, a portable and low-cost snapshot hyperspectral imaging (HSI) algorithm-based housing module for differentiating between original and duplicate holograms was proposed. The module consisted of a Raspberry Pi 4 processor, a Raspberry Pi camera, a display, and a light-emitting diode lighting system with a dimmer. A visible HSI algorithm that could convert an RGB image captured by the Raspberry Pi camera into a hyperspectral image was established. A specific region of interest was selected from the spectral image and mean gray value (MGV) and reflectivity were measured. Results suggested that shorter wavelengths are the most suitable for differentiating holograms when using MGV as the parameter for classification, while longer wavelengths are the most suitable when using reflectivity. The key features of this design include low cost, simplicity, lack of moving parts, and no requirement for an additional decoding key.

Detecting Asymptomatic Infections of Rice Bacterial Leaf Blight Using Hyperspectral Imaging and 3-Dimensional Convolutional Neural Network With Spectral Dilated Convolution

Rice is one of the most important food crops for human beings. Its total production ranks third in the grain crop output. Bacterial Leaf Blight (BLB), as one of the three major diseases of rice, occurs every year, posing a huge threat to rice production and safety. There is an asymptomatic period between the infection and the onset periods, and BLB will spread rapidly and widely under suitable conditions. Therefore, accurate detection of early asymptomatic BLB is very necessary. The purpose of this study was to test the feasibility of detecting early asymptomatic infection of the rice BLB disease based on hyperspectral imaging and Spectral Dilated Convolution 3-Dimensional Convolutional Neural Network (SDC-3DCNN). First, hyperspectral images were obtained from rice leaves infected with the BLB disease at the tillering stage. The spectrum was smoothed by the Savitzky-Golay (SG) method, and the wavelength between 450 and 950 nm was intercepted for analysis. Then Principal Component Analysis (PCA) and Random Forest (RF) were used to extract the feature information from the original spectra as inputs. The overall performance of the SDC-3DCNN model with different numbers of input features and different spectral dilated ratios was evaluated. Lastly, the saliency map visualization was used to explain the sensitivity of individual wavelengths. The results showed that the performance of the SDC-3DCNN model reached an accuracy of 95.4427% when the number of inputs is 50 characteristic wavelengths (extracted by RF) and the dilated ratio is set at 5. The saliency-sensitive wavelengths were identified in the range from 530 to 570 nm, which overlaps with the important wavelengths extracted by RF. According to our findings, combining hyperspectral imaging and deep learning can be a reliable approach for identifying early asymptomatic infection of the rice BLB disease, providing sufficient support for early warning and rice disease prevention.

Machine Learning Approaches in Traditional Chinese Medicine: A Systematic Review

Machine learning (ML), as a branch of artificial intelligence, acquires the potential and meaningful rules from the mass of data via diverse algorithms. Owing to all research of traditional Chinese medicine (TCM) belonging to the digitalization of clinical records or experimental works, a massive and complex amount of data has become an inextricable part of the related studies. It is thus not surprising that ML approaches, as novel and efficient tools to mine the useful knowledge from data, have created inroads in a diversity of scopes of TCM over the past decade of years. However, by browsing lots of literature, we find that not all of the ML approaches perform well in the same field. Upon further consideration, we infer that the specificity may inhere between the ML approaches and their applied fields. This systematic review focuses its attention on the four categories of ML approaches and their eight application scopes in TCM. According to the function, ML approaches are classified into four categories, including classification, regression, clustering, and dimensionality reduction, and into 14 models as follows in more detail: support vector machine, least square-support vector machine, logistic regression, partial least squares regression, k-means clustering, hierarchical cluster analysis, artificial neural network, back propagation neural network, convolutional neural network, decision tree, random forest, principal component analysis, partial least squares-discriminant analysis, and orthogonal partial least squares-discriminant analysis. The eight common applied fields are divided into two parts: one for TCM, such as the diagnosis of diseases, the determination of syndromes, and the analysis of prescription, and the other for the related researches of Chinese herbal medicine, such as the quality control, the identification of geographic origins, the pharmacodynamic material basis, the medicinal properties, and the pharmacokinetics and pharmacodynamics. Additionally, this paper discusses the function and feature difference among ML approaches when they are applied to the corresponding fields via comparing their principles. The specificity of each approach to its applied fields has also been affirmed, whereby laying a foundation for subsequent studies applying ML approaches to TCM.

Nondestructive Detection of Codling Moth Infestation in Apples Using Pixel-Based NIR Hyperspectral Imaging with Machine Learning and Feature Selection

Codling moth (CM) (Cydia pomonella L.), a devastating pest, creates a serious issue for apple production and marketing in apple-producing countries. Therefore, effective nondestructive early detection of external and internal defects in CM-infested apples could remarkably prevent postharvest losses and improve the quality of the final product. In this study, near-infrared (NIR) hyperspectral reflectance imaging in the wavelength range of 900–1700 nm was applied to detect CM infestation at the pixel level for three organic apple cultivars, namely Gala, Fuji and Granny Smith. An effective region of interest (ROI) acquisition procedure along with different machine learning and data processing methods were used to build robust and high accuracy classification models. Optimal wavelength selection was implemented using sequential stepwise selection methods to build multispectral imaging models for fast and effective classification purposes. The results showed that the infested and healthy samples were classified at pixel level with up to 97.4% total accuracy for validation dataset using a gradient tree boosting (GTB) ensemble classifier, among others. The feature selection algorithm obtained a maximum accuracy of 91.6% with only 22 selected wavelengths. These findings indicate the high potential of NIR hyperspectral imaging (HSI) in detecting and classifying latent CM infestation in apples of different cultivars.

Applications of deep-learning approaches in horticultural research: a review

Deep learning is known as a promising multifunctional tool for processing images and other big data. By assimilating large amounts of heterogeneous data, deep-learning technology provides reliable prediction results for complex and uncertain phenomena. Recently, it has been increasingly used by horticultural researchers to make sense of the large datasets produced during planting and postharvest processes. In this paper, we provided a brief introduction to deep-learning approaches and reviewed 71 recent research works in which deep-learning technologies were applied in the horticultural domain for variety recognition, yield estimation, quality detection, stress phenotyping detection, growth monitoring, and other tasks. We described in detail the application scenarios reported in the relevant literature, along with the applied models and frameworks, the used data, and the overall performance results. Finally, we discussed the current challenges and future trends of deep learning in horticultural research. The aim of this review is to assist researchers and provide guidance for them to fully understand the strengths and possible weaknesses when applying deep learning in horticultural sectors. We also hope that this review will encourage researchers to explore some significant examples of deep learning in horticultural science and will promote the advancement of intelligent horticulture.

Recent progress of nondestructive techniques for fruits damage inspection: a review

In the process of growing, harvesting, and storage, fruits are vulnerable to mechanical damage, microbial infections, and other types of damage, which not only reduce the quality of fruits, increase the risk of fungal infections, in turn greatly affect food safety, but also sharply reduce economic benefits. Hence, it is essential to identify damaged fruits in time. Rapid and nondestructive detection of fruits damage is in great demand. In this paper, the latest research progresses on the detection of fruits damage by nondestructive techniques, including visible/near-infrared spectroscopy, chlorophyll fluorescence techniques, computer vision, multispectral and hyperspectral imaging, structured-illumination reflectance imaging, laser-induced backscattering imaging, optical coherence tomography, nuclear magnetic resonance and imaging, X-ray imaging, electronic nose, thermography, and acoustic methods, are summarized. We briefly introduce the principles of these techniques, summarize their applicability. The challenges and future trends are also proposed to provide beneficial reference for future researches and real-world applications.

Optical Absorption and Scattering Properties at 900-1650 nm and Their Relationships with Soluble Solid Content and Soluble Sugars in Apple Flesh during Storage

Soluble solid content (SSC) is regarded as the most significant internal quality associated with the taste and maturity in fruits. Evaluating the relationship between the optical properties and soluble sugars facilitates exploration of the mechanism of optical techniques in SSC assessment. In this research, absorption coefficient (μ_a) and reduced scattering coefficient (μ′_s) of Fuji apple during storage were obtained using automatic integrating sphere (AIS) at 905–1650 nm. Relationships between μ_a, μ′_s and SSC, and soluble sugars contents, were investigated. The result showed that SSC, the content of total soluble sugars (TSS), fructose, glucose and sucrose were all decreasing after storage, and the same trend appeared in the change of μ_a and μ′_s. In the whole wavelength range, both μ_a and μ′_s were positively related to SSC and soluble sugars contents. The correlations between μ_a and SSC, and soluble sugars contents, showed increasing tendencies with increasing wavelengths, while for μ′_s, correlations had the opposite trend. The strongest correlations between μ_a and SSC, and soluble sugars contents, were observed in the correlation of μ_a with sucrose, with an r of 0.934. Furthermore, a partial least square (PLS) model for sucrose based on μ_a was built with the coefficient of determination of prediction (R_p²) and the root mean square error of prediction (RMSEP) of 0.851 and 1.047, respectively. The overall results demonstrate that optical properties at the range of 905–1650 nm could be used to evaluate SSC of apples and this may due to the strong correlation between sucrose content and μ_a.

Remote Sensing of Diseases

Detection, identification, and quantification of plant diseases by sensor techniques are expected to enable a more precise disease control, as sensors are sensitive, objective, and highly available for disease assessment. Recent progress in sensor technology and data processing is very promising; nevertheless, technical constraints and issues inherent to variability in host-pathogen interactions currently limit the use of sensors in various fields of application. The information from spectral [e.g., RGB (red, green, blue)], multispectral, and hyperspectral sensors that measure reflectance, fluorescence, and emission of radiation or from electronic noses that detect volatile organic compounds released from plants or pathogens, as well as the potential of sensors to characterize the health status of crops, is evaluated based on the recent literature. Phytopathological aspects of remote sensing of plant diseases across different scales and for various purposes are discussed, including spatial disease patterns, epidemic spread of pathogens, crop characteristics, and links to disease control. Future challenges in sensor use are identified.

Research Progress on the Early Monitoring of Pine Wilt Disease Using Hyperspectral Techniques

Pine wilt disease (PWD) caused by pine wood nematode (PWN, Bursaphelenchus xylophilus) originated in North America and has since spread to Asia and Europe. PWN is currently a quarantine object in 52 countries. In recent years, pine wilt disease has caused considerable economic losses to the pine forest production industry in China, as it is difficult to control. Thus, one of the key strategies for controlling pine wilt disease is to identify epidemic points as early as possible. The use of hyperspectral cameras mounted on drones is expected to enable PWD monitoring over large areas of forest, and hyperspectral images can reflect different stages of PWD. The trend of applying hyperspectral techniques to the monitoring of pine wilt disease is analyzed, and the corresponding strategies to address the existing technical problems are proposed, such as data collection of early warning stages, needs of using unmanned aerial vehicles (UAVs), and establishment of models after preprocessing.

A Hyperspectral Imaging Approach for Classifying Geographical Origins of Rhizoma Atractylodis Macrocephalae Using the Fusion of Spectrum-Image in VNIR and SWIR Ranges (VNIR-SWIR-FuSI)

Hyperspectral data processing technique has gained increasing interests in the field of chemical and biomedical analysis. However, appropriate approaches to fusing features of hyperspectral data-cube are still lacking. In this paper, a new data fusion approach was proposed and applied to discriminate Rhizoma Atractylodis Macrocephalae (RAM) slices from different geographical origins using hyperspectral imaging. Spectral and image features were extracted from hyperspectral data in visible and near-infrared (VNIR, 435-1042 nm) and short-wave infrared (SWIR, 898-1751 nm) ranges, respectively. Effective wavelengths were extracted from pre-processed spectral data by successive projection algorithm (SPA). Meanwhile, gray-level co-occurrence matrix (GLCM) and gray-level run-length matrix (GLRLM) were employed to extract textural variables. The fusion of spectrum-image in VNIR and SWIR ranges (VNIR-SWIR-FuSI) was implemented to integrate those features on three fusion dimensions, i.e., VNIR and SWIR fusion, spectrum and image fusion, and all data fusion. Based on data fusion, partial least squares-discriminant analysis (PLS-DA) and support vector machine (SVM) were utilized to establish calibration models. The results demonstrated that VNIR-SWIR-FuSI could achieve the best accuracies on both full bands (97.3%) and SPA bands (93.2%). In particular, VNIR-SWIR-FuSI on SPA bands achieved a classification accuracy of 93.2% with only 23 bands, which was significantly better than those based on spectra (80.9%) or images (79.7%). Thus it is more rapid and possible for industry applications. The current study demonstrated that hyperspectral imaging technique with data fusion holds the potential for rapid and nondestructive sorting of traditional Chinese medicines (TCMs).