Search

Organic acids are important components in determining fruit taste, and it is necessary to develop rapid analysis methods for them. A simple and effective method based on laser-engraved free-standing terahertz (THz) metamaterials has been developed for rapid analysis of fruit acids. Compared with traditional fabrication methods for THz metamaterials, laser direct writing technology was employed for fabricating the Cu-foil-based THz devices in this research, which is time-saving, cost-effective, and high-throughput. To realize effective fabrication of metamaterials, proper fabrication parameters were selected according to morphology and spectral characterization. Numerical analysis was performed to reveal the fundamental mechanism of THz sensing based on metamaterials. Furthermore, experimental results indicate that fruit acid with a concentration of 0.1 mg L−1 can be successfully distinguished, which is quick and simple compared with traditional acid–base titration. The proposed strategy demonstrates great potential for rapid in situ analysis of flavor components in food products.

Establishing a sustainable energy supply is necessary for intelligent greenhouse environmental management. Compared with traditional energy, green and eco-friendly energy is more conducive to protecting the agricultural production environment. In this study, a fluorinated superhydrophobic greenhouse film is proposed as a negative triboelectric layer material for the construction of a triboelectric nanogenerator that harvests raindrop energy (RDE-TENG). Moreover, an upgraded configuration is adopted, where the bulk effect between the lower/upper electrode and film replaces the interfacial effect of the liquid-solid interface, thereby promoting charge transfer. The results show that the RDE-TENG can serve as a sustainable energy source for greenhouse temperature and humidity sensors that assists in realizing intelligent control of the environment and guides agricultural production processes. This device exhibits high-voltage and a stable output; thus, it has the potential to replace traditional energy sources, which helps toward realizing a self-powered intelligent greenhouse planting mode.

The era of the Internet of Things (IoT) requires sustainable and convenient methods to power widely distributed sensing devices. Self-powered systems have emerged as a potential solution that utilizes ambient energy from environmental sources such as electromagnetic fields, mechanical motion, solar power, and temperature gradients. Recently, the integration of wireless technologies with self-powered systems has attracted significant attention as a way to address challenges in energy harvesting and transport without the cost and inherent physical constraints of wires. This review summarizes recent progress in the application of wireless technology in self-powered systems for applications in harvesting ambient electromagnetic energy and in transferring power between devices. In addition, challenges and development trends in the future of wireless self-powered sensor networks are discussed.

Background Rapid and non-destructive infrared spectroscopy has been applied to both internal and external quality evaluations of food and agro-products. Various linear and nonlinear chemometric methods have been developed for spectral analysis. The generalizability of previous chemometric methods is hindered by changing noise under various detection conditions and biological variabilities. Recently, deep learning approaches have been developed for spectral noise reduction, feature extraction, and calibration regression modeling. Scope and approach This review discusses the current challenges of conventional chemometric methods and the emerging deep learning approach for spectral analysis. The current state-of-the-art techniques, including unsupervised feature extraction and noise reduction models and supervised multivariate regression approaches, have been addressed in this review. The research on exploring the learning mechanism of the ‘black box’ deep learning model is also discussed. This review focuses on the application of deep learning approaches on quality evaluation of food and agro-products, lessons from current studies, and future perspectives. Key findings and conclusions The deep learning approach combined with spectroscopic sensing techniques has shown great potential for quality evaluation of food and agro-products. Current advances in deep learning-based qualitative analysis include variety identification, geographical origin detection, adulteration recognition, and bruise detection, whereas quantitative analysis includes multiple component content prediction for fruits, grains, and crops. The main advantage of deep learning approach is the decreasing the dependence on human domain knowledge by end-to-end analysis and the improved precision and generalizability.

Brazilian grain production increased more than fourfold from 1980 to 2016. The grain boom was achieved primarily by soybean–corn double cropping and cropland expansion—both show changing spatiotemporal patterns since the 1980s. Here, we quantified the contributions of these two strategies to corn and soybean production in Brazil using municipality-level data from 1980 to 2016. We found the contribution of double cropping to the grain boom steadily increased to 35% and the largest driving force was the increasing demand for grain export. While double cropping dominated the conventional agricultural regions, cropland expansion was still the major strategy in agricultural frontiers such as the Centre-West and Matopiba. The implementation of double cropping offset the equivalent of 76.7 million ha of Brazilian arable land for grain production from 2003 to 2016. Double cropping in Brazil has the potential to help alleviate land burdens in other pantropical countries with increasing global food demand. Soybean and corn production in Brazil increased more than fourfold from 1980 to 2016, which was achieved by double cropping and cropland expansion. The contribution patterns of the two strategies were spatiotemporally specific, while double cropping shows increasing impacts over land expansion to the grain boom.

Over the past few decades, optical waveguides have been increasingly used in wearable/implantable devices for on-body sensing. However, conventional optical waveguides are stiff, rigid, and brittle. A mismatch between conventional optical waveguides and complex biointerfaces makes wearable/implantable devices uncomfortable to wear and potentially unsafe. Soft and stretchable polymer optical waveguides not only inherit many advantages of conventional optical waveguides (e.g., immunity to electromagnetic interference and without electrical hazards) but also provide a new perspective for solving the mismatch between conventional optical waveguides and complex biointerfaces, which is essential for the development of light-based wearable/implantable sensors. In this review, polymer optical waveguides' unique properties, including flexibility, biocompatibility and biodegradability, porosity, and stimulus responsiveness, and their applications in the wearable/implantable field in recent years are summarized. Then, we briefly discuss the current challenges of high optical loss, unstable signal transmission, low manufacturing efficiency, and difficulty in deployment during implantation of flexible polymer optical waveguides, and propose some possible solutions to these problems.

Due to the unique characteristics of terahertz (THz) waves, such as low-energy, penetrating, and broadband, THz imaging has been regarded as cutting-edge technology in many fields. As the demand fo...

Wearable pressure sensors have gained widespread popularity. However, they fail to meet the demand of multi-scenarios applications and large-scale fabrication. Herein, we report a flexible, fully integrated, and wireless pressure sensing chip system by integrating a sensitivity-tunable sensor with a flexible printed circuit board on polyimide films. The sensor incorporates laser-induced graphene-based interdigitated electrode with a conductive sponge fabricated by in situ growth of silver nanoparticles (AgNPs) on a MXene-coated sponge. The sensitivity can be tuned by changing the growth density of AgNPs, with a higher density of AgNPs corresponding to higher sensitivity. The highest sensitivity can reach 0.90 kPa−1, which is higher than the sensor without AgNPs (0.10 kPa−1). This system is expected to find applications in multi-scenarios, such as fruit growth and pulse signal. All these studies indicate that this integrated sensing chip system possesses the potential in improving the fabrication efficiency and sensing convenience for future wearable electronics.

Plants have complex internal signaling pathways to quickly adjust to environmental changes and harvest energy from the environment. Facing the growing population, there is an urgent need for plant transformation and precise monitoring of plant growth to improve crop yields. Nanotechnology, an interdisciplinary research field, has recently been boosting plant yields and meeting global energy needs. In this context, a new field, “plant nanoscience,” which describes the interaction between plants and nanotechnology, emerges as the times require. Nanosensors, nanofertilizers, nanopesticides, and nano‐plant genetic engineering are of great help in increasing crop yields. Nanogenerators are helping to develop the potential of plants in the field of energy harvesting. Furthermore, the uptake and internalization of nanomaterials in plants and the possible effects are also worthy of attention. In this review, a forward‐looking perspective on the plant nanoscience is presented and feasible solutions for future food shortages and energy crises are provided., This review provides the first detailed introduction of plant nanoscience, including nanosensors, nanofertilizers, nanopesticides, nanomaterials‐mediated gene transformation, nanogenerators, and nanotoxicology. The interaction between nanomaterials and plants, as well as, the advanced application fields of nanotechnology in plants are discussed deeply to better manage the growth of plants and offer guidance for the future research and practical applications.

Non-destructive detection methods to identify food foreign contaminants meet the development of the food industry. Near infrared (NIR) spectroscopy and computer vision (CV) have been considerably intriguing due to the advantages of safety and rapidity. In this study, foreign contaminants, including metallic iron, polypropylene plastic, and hair in bread, were identified based on these two techniques. In NIR spectroscopy, the effectiveness of distance match and discriminant analysis methods combined with different spectral pretreatments was compared. It showed that the accuracy was 98%, 94%, 91%, and F-score was 0.97, 0.93, and 0.91 in the validation set for detecting foreign contaminants aforementioned when discriminant analysis combined with Savitzky-Golay smoothing was used. In CV, deep learning based on the modified U-net has applied to segment these contaminants on the surface of the bread; the accuracy of the test set was 95%, 93%, and 92%, respectively. Based on the results, it can be concluded that NIR spectroscopy and CV are both an operational way to detect foreign contaminants in bread. And these two techniques could be combined to apply in on-line detection further.

Terahertz (THz) electromagnetic waves strongly interact with complex molecules, making THz spectroscopy a promising tool for high-sensitivity molecular detection, especially for biomedical applications. Metamaterials are typically used for enhancing THz-molecule interactions to achieve higher sensitivities. However, a primary challenge in THz molecular sensing based on metallic metamaterials is the limited tunability of optical constants of metals. Here, we present an ultrahigh-sensitivity molecular sensor based on carbon nanotube (CNT) THz metamaterials. The sensor, consisting of a CNT cut-wire array on a Si substrate prepared by a novel two-step method, exhibits a reflectance resonance whose frequency strongly varies with the substrate composition, geometries of periodic arrays, and analyte composition. We used this sensor to detect glucose, lactose, and chlorpyrifos-methyl molecules, achieving limit-of-detection values of 30, 40, and 10 ng/mL (S/N = 3), respectively, higher than that of metallic metamaterials by 2 orders of magnitude. We attribute this ultrahigh sensitivity to the high conductivity of CNTs and the efficient adsorption of the target analyte by CNTs through van der Waals forces and π-π stacking. These easy-to-fabricate CNT-based THz metamaterials pave the way for versatile and reliable ultrahigh-sensitivity THz molecular detection.

Background Monte Carlo (MC) has been widely used in fields such as biomedicine and computer graphics owing to its unique capabilities of flexibility, high-accuracy and simplicity for modeling light transport in tissues, but the applications in food and agricultural domain are limited or hindered due to the lack of knowledge on the optical properties of food products. Thanks to major breakthroughs in optical measuring and computing technologies since the year of 2000, significant advances have been made in sensing techniques for measuring tissue optical properties. Therefore, MC has witnessed great progress in food and agricultural domain over the past two decades. Scope and approach The development of MC for modeling light transport in food and agricultural products, including the principle, advanced MC methods, relevant applications, and future perspectives were reviewed. The paper is aimed at helping interested researchers to gain a better understanding of the MC technique, thus stimulating quality and safety assessment of food and agricultural products. Key findings and conclusions This paper provides an overview of the procedure of MC modeling for light transport in food and agricultural products and commonly used MC models. Advanced methods for accelerating MC simulations are then presented. Applications of MC simulations in food and agricultural products, since the year of 2000, for optimizing the design of sensing configuration and parameter, estimating tissue optical property, and assessing quality and safety are then reviewed. Finally, challenges and future perspectives for MC technique in modeling light transport are discussed.

Electrolyte-gated transistors (EGTs) hold great promise for next-generation printed logic circuitry, biocompatible integrated sensors, and neuromorphic devices. However, EGT-based complementary circuits with high voltage gain and ultralow driving voltage (110) under a supply voltage of only 0.7 V. Furthermore, NAND and NOR logic circuits on both rigid and flexible substrates are realized, enabling not only excellent logic response with driving voltages as low as 0.2 V but also impressive mechanical flexibility down to 1-mm bending radii. Finally, the HCIN was applied in electrooculographic (EOG) signal monitoring for recording eye movement, which is critical for the development of wearable medical sensors and also interfaces for human–computer interaction; the high voltage amplification of the present HCIN enables EOG signal amplification and monitoring in which a small ∼1.5 mV signal is amplified to ∼30 mV.

Terahertz (THz) spectroscopy for highly sensitive and selective detection has attracted an increasing recognition. To improve the sensing performance, THz metamaterials were served as signal enhancing devices. However, the interaction between molecules and bulk materials (such as metal and silicon, frequently used in fabricating metamaterials) is simple, which limits the sensing performance. Here we introduce 2D materials into THz sensors, respectively. Through π-π stacking and Van der Waal's force, such 2D materials integrated absorbers show high molecule detection and recognition ability. Also the flexible sensor present strong robustness over bending. The results show that our 2D materials based sensors are flexible, stable, and versatile, suited for a broad range of sensing applications.

The boom of plant phenotype highlights the need to measure the physiological characteristics of an individual plant. However, continuous real‐time monitoring of a plant's internal physiological status remains challenging using traditional silicon‐based sensor technology, due to the fundamental mismatch between rigid sensors and soft and curved plant surfaces. Here, the first flexible electronic sensing device is reported that can harmlessly cohabitate with the plant and continuously monitor its stem sap flow, a critical plant physiological characteristic for analyzing plant health, water consumption, and nutrient distribution. Due to a special design and the materials chosen, the realized plant‐wearable sensor is thin, soft, lightweight, air/water/light‐permeable, and shows excellent biocompatibility, therefore enabling the sap flow detection in a continuous and non‐destructive manner. The sensor can serve as a noninvasive, high‐throughput, low‐cost toolbox, and holds excellent potentials in phenotyping. Furthermore, the real‐time investigation on stem flow insides watermelon reveals a previously unknown day/night shift pattern of water allocation between fruit and its adjacent branch, which has not been reported before., A plant‐wearable sensor in situ reveals water transport inside a plant. Due to a special design and the materials chosen, the realized plant‐wearable sensor is thin, soft, lightweight, air/water/light‐permeable, and shows excellent biocompatibility, therefore enabling detection in a continuous and non‐destructive manner.

The growing demand of intelligent wearable devices has triggered rapid development of highly stretchable strain sensors and power sources. Here, we fabricated a sandwich-structured stretchable device, where a composite film is encapsulated in two layers of polydimethylsiloxane (PDMS). The highly conductive composite film is composed of one-dimensional (1D) silver nanowires (AgNWs) network wrapped with two-dimensional (2D) metallic MoS2 nanosheets. The combination of 1D AgNWs and 2D metallic MoS2 nanosheets endows the composite film with excellent flexibility and conductivity, which enables their use as high-performance skin-mountable strain sensor with the maximum gauge factor of 215.4 and high stretchability up to 70%. Furthermore, the sandwich-structured device can serve as a stretchable triboelectric nanogenerator (STENG). Results show that the STENG with an area of 1 × 2.5 cm2 could produce an average power density of 0.16 W/m2. In addition, the STENG could retain stable output performance even under a stretching ratio of 50%. With the outstanding stretchability, it can be fully conformal on nonplanar irregular object such as plant leaves. Moreover, we demonstrated that the fabricated STENG could harvest wind energy and serve as a self-powered wind speed sensor when attached to plant leaves. This work provides new prospects for stretchable power sources and shows potential applications in wearable electronics.

In this work, a fully stretchable triboelectric nanogenerator (FSTENG) composed of an electrospun electrode and a porous PDMS friction layer with anti-nickel foam structure was fabricated. This FSTENG can deliver an optimal output voltage of 92 V, which is ten times greater than that of the TENG based on flat PDMS film. The superior performance is attributed to the use of anti-nickel foam structure-based porous PDMS film in FSTENG, which enhances the surface-to-volume ratio and surface roughness. Importantly, the FSTENG could maintain stable output performance even under a stretching ratio of 50%. With these excellent features, the developed FSTENG was designed as a self-driven sensor for finger bending and walking status monitoring. Furthermore, to expand the driving mode of FSTENG, FSTENG was successfully attached to the leaf for monitoring the wind speed in the natural environment without external power supply. The excellent flexibility, stretchability, and versatility of FSTENG make it as a promising self-driven sensor for human behavior and environmental monitoring.

Detection of foreign objects is crucial for quantitative image analysis in numerous food quality and safety inspection applications. Rapid detection of foreign objects in walnuts using computer vision still faces challenge due to the irregular shapes and complex features of foreign objects. Some detection methods require application specific transforms, expertly designed constraints and model parameters, and have limited detection performance due to their maintenance costs. In recent years, deep learning has become a focus in different research fields, because methods based on deep learning are able to directly learn features from training data. In this study, we apply two different convolutional neural network structures to walnut images for automatically segmenting images and detecting different-sized natural foreign objects (e.g., flesh leaf debris, dried leaf debris and gravel dust) and man-made foreign objects (e.g., paper scraps, packing material, plastic scraps and metal parts). The proposed deep-learning method is simpler because it avoids extracting features manually, and overcomes the conglomeration phenomenon between walnuts and foreign objects in actual images. The proposed method is able to correctly segment 99.5% of the object regions in the 101 test images and to correctly classify 95% of the foreign objects in the 277 validation images. The segmentation and detection processing time of each image was less than 50 ms. Future work will focus on deep learning using multi-waveband imaging hardware and fast on-line inspection control for the equipment and robots.

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.

Over the past decade, a wide range of nanomaterials have attracted extensive attention for artificial enzymes (nanozymes) to meet task-specific applications because of their intrinsic enzyme-like catalytic activities. Artificial enzymes based on inorganic nanomaterials are now envisioned as an increasingly relevant alternative to traditional natural enzymes due to their structure robustness, unusual high catalytic performance, facile synthesis from inexpensive and abundant raw materials. This tutorial review summarizes and highlights latest progress in design and construction strategies of inorganic nanomaterials to mimic the structure and functionality of natural peroxidase with a focus on optical technique, both the existing research hotspots and future prospects are outlined and discussed. Furthermore, methodologies for tailoring nanomaterials-based intrinsic peroxidase mimics to enhance their catalytic activity including morphology control, surface modulation and heterostructure coupling are also summarized. Finally, we put forward our personal insights into future research challenges and opportunities, hoping to motivate more researchers to participate in the research in this emerging field.

The accurate and rapid detection of foreign materials in food products is essential for ensuring food safety and quality. Terahertz (THz) imaging is an emerging technology for non-destructive detection in food industrial with the advantages of non-ionization and spectroscopic fingerprinting. In this study, we have attempted to explore the effectiveness of THz spectroscopic imaging for foreign material detection in food with complicated compositions. The feasibility of locating contaminations in sausages were demonstrated through typical spectral comparison, spectral classification assisted by principal component analysis and discriminant analysis methods, and one-dimensional scanning spectral analysis. Our imaging results indicated THz spectroscopic imaging has the ability to locate metal contaminations in sausages no matter what part of sausages is as the food matrix. This study will provide new knowledge about foreign material detection in food with complicated compositions and a basis for parameters selection of continuous-wave THz imaging for contamination detection in this case.

Terahertz (THz) radiation attracted great interest in the fields of material characterization, nondestructive security screening, clinical diagnostics, and identification of chemicals and molecules. Label-free THz sensing of trace amount of targets including biomolecules is promising because of their rich spectral fingerprint in this electromagnetic region; however, improving the sensitivity remains to be a challenge, partially due to the limitations of THz sources and detectors. The resonantly enhanced electromagnetic fields in metamaterials and metasurfaces offer a potentially viable solution, although highly complicated decoration process is still needed for biosensing on the surface of metamaterials. Here we demonstrate a simple biosensing platform by integrating a monolayer graphene on a THz metamaterial absorber cavity, where the introduction of sensing targets results in a large change of the metamaterial resonant absorption (or reflection) because of their strong interaction with graphene. We experimentally show its ultrahigh sensitivity through detecting trace amount of chlorpyrifos methyl down to 0.2 ng. Using simple decoration steps and utilizing DNA to capture thrombin, we further show the feasibility of this platform serving as a sensitive biosensor. Accepted version The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (No. 31471410) and Research Fund for the Doctoral Program of Higher Education of China (No. 20130101110036). H.T.C acknowledges support in part from LANL LDRD Program and the Center for Integrated Nanotechnologies.

A graphene paper-based flexible solid-contact ion-selective electrode (SC-ISE) was developed to detect lead ion sensitively. Graphene paper obtained via a simple vacuum filtration method was used as the electrode substrate for direct coating of an ion-selective membrane. The Nernstian slope of the prepared paper-based potentiometric sensor toward lead ion detection was demonstrated as 29.4 mV per decade. A detection limit as low as 2.5 × 10-7 mol L-1 was achieved. Reversed chronopotentiometry and water layer test revealed that the graphene paper-based SC-ISE possessed excellent potential stability because of the hydrophobicity of graphene paper. Furthermore, reliable data were obtained from the detection of lead ion levels in real water samples using the graphene paper-based potentiometric sensor, which shows great potential in practical application. Keywords: Graphene paper, Heavy metal, Ion-selective electrode, Potentiometry, Water sample.

Perching is a natural behavior of poultry. Considerable research has been done to explore the relationship between group overall perch usage and well-being of laying hens. To quantify the potential cause-effect relationship on individual hens with different health or well-being status (e.g., keel bone deformation, foot pad lesion, social ranking) in a group, it is necessary to identify the perching behavior of individual birds. However, continuously monitoring individual birds in a group poses considerable challenges. To enable such research and potential commercial application, this study developed and validated a radio frequency identification (RFID) based automated perching monitoring system (APMS) for characterizing individual perching behaviors of group-housed poultry. The APMS consisted of an RFID module, a load cell module, and a round wooden perch. The RFID module was comprised of a high-frequency RFID reader, three customized rectangular antennas placed under the perch, and RFID transponders attached to the birds. The load cell module was comprised of a data acquisition system and two load cells supporting both ends of the perch. The daily number of perch visits (PV) and perching duration (PD) for individual birds were used to delineate perching behavior. Three identical experimental pens, five hens per pen, were equipped with the monitoring system. Two RFID transponders were attached to each hen (one per leg), and a distinct color was marked on the bird’s head for video or visual identification and validation. Performance of the APMS was validated by comparing the system outputs with manual observation and labeling over an entire day. Sensitivity and specificity of the system were shown to improve from 97.77% and 99.88%, respectively when using only the RFID module to 99.83% and 99.93% when incorporating weight information from the load cell module. Using this system, we conducted a preliminary trial on the relationship of perching behavior and body weight of laying hens, which revealed little effect of body weight but considerable variability in perching behavior among the individual hens. The study demonstrated that the APMS had excellent performance in measuring perching behaviors of individual birds in a group. The APMS offers great potential for delineating individual differences in perching behavior among hens with different social status or health conditions in a group setting. Keywords: Individual perching behavior, Laying hen, Load cell, Precision livestock farming, RFID, Welfare.

Real-time monitoring of the wind vector, including wind speed and direction, is significant for crop management and production in agricultural environment. However, traditional wind vector sensors are difficult to meet the needs of modern sustainable agriculture due to their considerable size, low accuracy, and lack of energy supply. Hence, there is a great need for wind vector sensors that can respond super-sensitively to environmental stimuli without external power supply. Herein, a transparent and degradable hydroxyethyl cellulose film was introduced to fabricate a triboelectric nanogenerator (TENG) based on the slit effect for wind energy harvesting and self-powered wind vector monitoring. The maximum output voltage of slit-effect-enabled TENG (SE-TENG) is up to ~ 600 V under a wind speed of 7 m/s, and various agricultural sensors can be continuously driven. Furthermore, an omnidirectional wind energy harvester (OWEH) composed of eight SE-TENGs and an electrometer data-acquisition board was assembled to monitor the wind directions and speed ranging from 0.5 to 10 m/s accurately. This work poses an effective and green method for wind vector monitoring in agricultural environment and demonstrates the potential of TENG technology in intelligent agriculture.

Spectroscopic techniques have been widely applied in agricultural applications. The development of calibration transfer is promising for the robust analysis of spectral data collected by varying instruments. The reliance on standard samples for standardization approaches remains a critical challenge for on-site applications. In this study, a deep learning approach, named DeepTranSpectra (DTS), is proposed to transfer convolutional neural network models among multiple near-infrared spectrometers with different types. The proposed DTS approach effectively avoids the requirements for standard samples by using labeled samples of slave instruments. The calibration transfer analysis is investigated on a soybean meal and one wheat dataset for predicting moisture and crude protein contents. The developed DTS approach demonstrates improved transfer performance compared with three popular standardization approaches, including piecewise direct standardization (PDS), canonical correlation analysis (CCA), and slope and bias correction (SBC). A feature visualization method is leveraged to interpret the transfer mechanism of the DTS approach. The interpretation results show that the DTS approach refines the model parameters to adapt to slave devices based on critical features of the master calibration. The DTS approach provides advanced reliability under different sample selections in Monte-Carlo cross-validation. The integration of deep learning approaches with calibration transfer analysis facilitates agricultural applications for emerging deep learning-based chemometric analysis.

There is an increasing recognition that terahertz (THz) spectroscopy can be used for high-sensitivity molecular sensing. Therefore, in recent years, much work has been devoted to developing flexible, compact, and high-sensitivity THz sensors. However, most designs employ metamaterials, which require complicated, and often expensive, fabrication procedures. Also, the metamaterial structures create a gap between the sensor surface and the target surface, which decreases the effective contact area between them, resulting in reduced sensing performance. Here, we fabricated a metamaterial-free graphene-based THz sensor with user-designed patterns for sensing at bio-interfaces. External molecules can strongly interact with π electrons in graphene, which moves the Fermi level and changes the amount of THz absorption. We used this sensor to successfully detect chlorpyrifos methyl with a limit of detection at 0.13 mg/L. We also detected pesticide molecules of a concentration of 0.60 mg/L on the surface of an apple, revealing the flexibility of this sensor. The flexible graphene THz sensor showed high sensing stability and robustness over 1000 cycles of bending. These results show that our graphene-based thin-film sensors are easy to fabricate, flexible, versatile, and suited for a wide range of sensing applications.

Terahertz (THz) technique has become one of the most promising analytical methods and has been applied in many fields. Attenuated total reflection (ATR) technique applied in THz spectroscopy and imaging has been proven to be superior in functionalities such as modulation, sensing, analyzing, and imaging. Here, we first provide a concise introduction to the principle of ATR, discuss the factors that impact the ATR system, and demonstrate recent advances on THz wave modulation and THz surface plasmon sensing based on the THz-ATR system. Then, applications on THz-ATR spectroscopy and imaging are reviewed. Towards the later part, the advantages and limitations of THz-ATR are summarized, and prospects of modulation, surface plasmon sensing, spectroscopy and imaging are discussed.

Potato tubers suffer from dynamic loads during series of harvest and/or post-harvest operations, leading to the occurrences of mechanical damages. Supervision of dynamic responses upon impact from varied drop heights (200, 400 and 600 mm) onto rigid plate for potatoes was undertaken using finite element analysis (FEA) in this study. Two image processing-aided algorithms were developed and validated with reference methods for creating a potato digital model for FEA and determining the radius of curvature on contact point of the potato. A potato was taken as elastic and viscoelastic models in FEA respectively. The material parameters were derived from compression test and stress-relaxation test. Five response parameters of contact time Tc, peak force time Tp, maximum impact force Fmax, maximum deformation Dmax and coefficient of restitution r were selected to characterize dynamic response of potato collision. Results showed that Tc and Tp decreased slightly with the rise of drop height, while Fmax and Dmax showed positive correlations with drop heights. The value of r stayed high despite of material model and drop height. Dynamic responses presented similar but different values for viscoelastic model compared with elastic model, due to its viscous effect. It was also found that simulation results were highly dependent on modulus of elasticity for both models. Conclusions from this study can be used for monitoring occurrences of damages and solving other agricultural engineering problems with respect to material impact/collision.

trans-Resveratrol (TRA), is one of the indicators to evaluate the quality of red wines. In this study, a novel flexible electrochemical sensor using direct laser-induced graphene (LIG) technique that transforms the commercial Kapton tape into three-dimensional (3D) porous graphene was developed for sensitive detection of TRA molecules in red wines. For the first time, the strategy of ‘double layer’ (Kapton/polyimide) was employed to obtain the LIG with undamaged shape and excellent electrical properties. The mechanism of heat absorption and dissipation of laser-induced process was investigated in detail. The prepared electrochemical sensor with excellent repeatability, stability, reproducibility, and reliability, appears an excellent linear response within the TRA concentration range from 0.2 to 50 μmol L−1 and a low limit of detection (LOD) of 0.16 μmol L−1. Furthermore, the developed sensor can be applied for the evaluation of TRA level in red wines and grape skins with a satisfactory result. The sensor may be potential in analysis of active compounds in food or environmental samples.

Deep learning approaches, especially convolutional neural network (CNN) models, have achieved excellent performances in vibrational spectral analysis. The critical drawback of the CNN approach is the lack of interpretation, and it is regarded as a black box. Interpreting the learning mechanism of chemometric models is critical for intuitive understanding and further application. In this study, an interpretable CNN model with a global average pooling layer is presented for Raman and mid-infrared spectral data analysis. A class activation mapping (CAM)-based approach is leveraged to visualize the active variables in the whole spectrum. The visualization of active variables shows a discriminative pattern in which the most contributed variables peaked around theoretical chemical characteristic bands. The visualization of the feature maps by three convolutional layers demonstrates the data transformation pipeline and how the CNN model hierarchically extracts informative spectral features. The first layer acts as a Savitzky-Golay filter and learns spectral shape characteristics, while the second layer learns enhanced patterns from typical spectral peaks on a few correlated variables. The third layer shows stable activations on critical spectral peaks. A partial least squares - linear discriminant analysis (PLS-LDA) model is presented for comparison on classification accuracy and model interpretation. The CNN model yields mean classification accuracies of 99.01 and 100% for E. coli and meat datasets on the test set, while the PLS-LDA models obtain accuracies of 98.83 and 100%. Both the CNN and PLS-LDA models demonstrate stable patterns on active variables while CNN models are more stable than PLS-LDA models on classification performances for various dataset partitions with Monte-Carlo cross-validation.

Two kinds of two-dimensional MXene (of type Ti3C2Tx and Ti2CTx) nanosheets are described for use in solid-contact ion-selective electrodes (SC-ISEs) where they act as ion-to-electron transducers. Electrochemical characterizations show that the MXene-coated electrodes possess high double layer capacitance and enable rapid electron transport. This demonstrates the enhanced efficiency of MXene-based solid-contact layers to improve ion-electron transduction. Both Ti3C2Tx- and Ti2CTx-based SC-ISEs exhibited a Nernstian response (26.4 and 24.9 mV/decade, respectively) between 10−1 and 10–5.5 M Ca(II) concentrations with rapid response (

Terahertz (THz) technology has been widely utilized in fields of multiple disciplines because of its unique properties such as broadband and directivity. With existing and emerging technologies increasing demand for compact, fast and broadband terahertz devices, high-performance THz devices are becoming indispensable. Carbon nanotubes (CNTs) with superior optical and electronic performance have prompted intense research and significant advances, paving the way for realistic applications. In this review, we introduce the mechanisms of utilizing carbon nanotubes to generate, modulate, polarize, and detect terahertz radiation. We also discuss recent carbon nanotube terahertz devices based on these mechanisms, including THz antennae, emitters, amplifiers, transistors, polarizers, and1 detectors. Furthermore, the main performance achievement for such devices is summarized briefly, and an outlook on the performance improvement for existing deficiencies is presented.

This research was aimed at optimizing the inverse algorithm for estimating the optical absorption (μa) and reduced scattering (μs′) coefficients from spatial frequency domain diffuse reflectance. Studies were first conducted to determine the optimal frequency resolution and start and end frequencies in terms of the reciprocal of mean free path (1/mfp′). The results showed that the optimal frequency resolution increased with μs′ and remained stable when μs′ was larger than 2 mm−1. The optimal end frequency decreased from 0.3/mfp′ to 0.16/mfp′ with μs′ ranging from 0.4 mm−1 to 3 mm−1, while the optimal start frequency remained at 0 mm−1. A two-step parameter estimation method was proposed based on the optimized frequency parameters, which improved estimation accuracies by 37.5% and 9.8% for μa and μs′, respectively, compared with the conventional one-step method. Experimental validations with seven liquid optical phantoms showed that the optimized algorithm resulted in the mean absolute errors of 15.4%, 7.6%, 5.0% for μa and 16.4%, 18.0%, 18.3% for μs′ at the wavelengths of 675 nm, 700 nm, and 715 nm, respectively. Hence, implementation of the optimized parameter estimation method should be considered in order to improve the measurement of optical properties of biological materials when using spatial frequency domain imaging technique.

Terahertz technology, including terahertz spectroscopy and terahertz imaging, is gaining increasing attention and plays an important role in various fields. With the development of terahertz sources and detectors in recent decades, terahertz imaging has been applied in many fields, including security screening, material evaluation, biomedicine, agriculture, and food science. However, progress has been comparatively slow in applying terahertz imaging to agriculture and food engineering. In this article, some commonly used terahertz imaging systems are summarized, and various applications of terahertz imaging in agriculture and food engineering are reviewed, including food safety and quality inspection, seed inspection, water content evaluation of plant leaves, and others. Challenges and the future outlook of terahertz imaging are also discussed. Keywords: Agriculture, Food, Instrumentation, Seed, Terahertz imaging.

Multi-temporal deep learning approaches have exhibited excellent classification performance in large-scale crop mapping. These approaches efficiently and automatically transform remote sensing time series into high-dimensional feature representations to identify crop types. The lack of interpretation, however, is regarded as a major drawback of these high-performance approaches. Interpreting deep learning approaches in multi-temporal crop mapping is critical for verifying their reliability. This study aims to quantify the impact of multi-temporal information in input time series on classification performance and develop a multi-perspective interpretation pipeline for deep learning models. The pipeline involves three interpretation approaches: evaluating input feature importance, analyzing hidden features, and monitoring temporal changes in model's soft output. An experiment is conducted to classify corn and soybean in the U.S corn belt in 2018. The study area consists of three sites each encompassing millions of pixel-level samples at 30 m resolution. The Landsat Analysis Ready Data are used as the input remote sensing time series and Cropland Data Layer is used as the ground reference. Attention-based Long Short-Term Memory (AtLSTM) and Transformer models are built as multi-temporal deep learning models, and compared to Random Forest (RF). Complete time series input in the correct order achieves a higher overall accuracy of 97.8% than using single-window or out-of-order inputs, indicating multi-temporal information facilitates crop classification. An assessment of the input feature importance demonstrates that the AtLSTM, Transformer, and RF models all consider the period from weeks 11 to 20 (early-July to late-August) as a key growth period and the shortwave infrared band as the critical band for corn and soybean discrimination. Hidden feature analysis suggests that the AtLSTM model accumulates the useful information over the growth period, while the Transformer model extracts the temporal dependencies that contribute important information to high-level feature learning. The learned features contain more effective and refined information than the raw input features and thus are better suited for crop classification. The soft output analysis in the in-season classification scenario demonstrates that increased length of input time series improves the model's confidence in the classification results. The further comparison of input feature importance in different sites and years demonstrates the applicability of the interpretation approach at larger spatiotemporal extents with heterogeneous landscapes and interannual variability. This study provides a multi-perspective evaluation to identify key features in multi-spectral and multi-temporal remote sensing data, and yields a practical approach to integrate agronomy knowledge in deep learning-based crop mapping.

Mycotoxins have been regarded as one of the most severe threatens to food safety and environment, however, conventional detection methods are limited by low collection efficiency and tedious detection procedures. In this work, adsorptive and responsive hybrid sponge was prepared by integrating porous fluorescent Zr-LMOF (luminescent metal-organic framework) and robust melamine sponge for simultaneous collection/removal and detection of mycotoxins. The hybrid sponge achieved high load capacity of 15.8 mg g−1 within 10 min and worked well in a wide range of pH and temperatures. The fluorescence of the hybrid sponge was efficiently quenched by AFB1 in the aqueous water in 10 min due to the hierarchical pore structure enabled absorbent-target interaction. This simple, robust and low-cost dual functional hybrid sponge enabled the detection with relatively low LOD of 1.6 ppb and wide linear range of 31.2 ppb-15.6 ppm. Additionally, AFB1 in water was quickly absorbed by the Zr-LMOF crystals on the melamine microfibers with the removal efficiency reached as high as 90%. The removal efficiency towards other mycotoxins including aflatoxin B2, aflatoxin G1, aflatoxin G2, aflatoxin M1 and ochratoxin A also reached 80%. Overall, this low-cost, robust and elastic hybrid sponge has great potential in the construction of novel sensors for practical applications to ensure environmental and agricultural safety.

Example of defect detection of on-line orange. Original RGB images (top), manually marked defects in blue (second row), defect region segmentation directly in uncorrected image (third row), manually marked defects in corrected image in blue by the proposed algorithm (fourth row), defects detection results (bottom).Display Omitted Fix uneven lightness distribution problem.Correction algorithm is adaptive in approach.Common defective orange detection at low cost. Automatic detection of defective fruit by computer vision system still faces challenge due to the uneven lightness distribution on the surface. A fast adaptive lightness correction algorithm implementation which is simpler and easier in real-time approach is proposed to overcome interferences from non-uniform reflectance intensity distribution on moving fruit surface and avoids error detection. The algorithm is tested by on-line and static defective orange images in the different lighting conditions. This study also compares other lightness correction algorithm implementations for defect detection. Recently, embedded vision systems are more and more popular because of low-cost, compact size and stability. A low-cost embedded vision system based on an industry gigabit ethernet camera and embedded Linux system in an arm processor with limited computing power compared with high-performance major PC is also originally developed to test and prove the performance of fast adaptive lightness correction algorithm using the most common surface defect of Navel orange in China. The time consumption of adaptive lightness correction algorithm of an image is below 6ms. The processing time of an orange image is below 30ms.

The gray-level image after removing background, (b) the binary image.Display Omitted Successful detections of various types of surface defects.Avoiding additionalimage lightness correction process.Image detection algorithm is novel and practical. Automatic detection of defective oranges by computer vision system is not easy because of the uneven lightness distribution on the surface of oranges. It means that the methods onlydirectly using global segmentation provide unsatisfactory results when orange images present faint defect characters or inhomogeneous surface. The contrast between sound and defective regions can be used to produce more accurate segmentation results, which is more capable of detecting pixels lying around the defect boundary on orange surface based on the local segmentation method. In this paper, we study and propose a sliding comparison window local segmentation algorithm and also presents the detailed image processing procedure including removal of background pixels, image binarization using local segmentation, image subtraction, image morphological modification, removal of stem end pixels for detecting surface defect in an orange gray-level image. This method is an original contribution that allows successful segmentation of various types of surface defects (e.g., insect injury, wind scarring, thrips scarring, scale infestation, canker spot, dehiscent fruit, copper burn, phytotoxicity).The image segmentation algorithm was tested with 1191 samples of oranges. The proposed algorithm was able to correctly detect 97% of the defective orange. Future work will be focused on whole surface and fast on-line inspection.

Spatially-resolved diffuse reflectance (SRDR) technique, with simple instrumentation, fast measurement, and relatively cheap devices, offers great potential for measurement of optical absorption (µa) and scattering (µs′) properties of agro-products. In this study, a laboratory-based double-fiber-optical-probe system was established and developed for optical property measurement of pear samples in the spectral region of 500–1000 nm. The system, based on a simple two-source diffusion theory model, was able to acquire SRDR profiles, which were used to calculate the µa and µs′ via nonlinear fitting. 32 pear samples with three different measured surfaces were tested in this study. The SRDR profiles of 41 positions were acquired, from −1 to 1 cm at an interval of 0.05 cm. The values of µa ranged nearly from 0.10 to 0.61 cm−1 and peaked at 970 nm due to water absorption, while the µs′ decreased along wavelengths with its values from 12.5 to 9.5 cm−1. The results indicated that the optical properties of pear tissues were wavelength-dependent, and small discrepancies of µa and µs′ could be found for different measured surfaces. The creation of standard sample slicing procedure is in great demand, which is useful for improving the accuracy of optical property measurement.

Rolling poultry eggs through a weighing rail is a simple, fast, and efficient way to measure mass. However, disturbances generated primarily by the rolling eggs vary with the rolling state and speed as well as with the egg mass and shape. These disturbances distort the load cell signal in the form of a stochastic strong amplitude fluctuation, and eliminating these fluctuations completely is a difficult task. A piezoelectric accelerometer was used to measure the vibration disturbances. The smoothed pseudo-Wigner-Ville distribution (SPWVD), which is a type of time-frequency analysis, was used to analyze the vibration disturbances using the accelerometer data. The results indicated that the disturbances were non-stationary, and the frequency characteristics were time-varying. Preliminary experiments showed that the commonly used low-pass filtering and subsequent average-based mass estimation method (AME) did not result in a satisfactory weighing accuracy. To meet the requirement of fast and accurate dynamic weighing of eggs, this research proposed a sorting-based mass estimator (SME) that consisted of an optimized digital filter and asymmetrically trimmed mean. The SME regarded the mass measurement as a problem of location estimation of non-Gaussian and heavy-tailed random variables, given the short observation time and the presence of outliers (disturbances). Four types of digital filters in the SME were selected to pre-filter the load cell data. The relevant parameters of the digital filters and asymmetrically trimmed mean in the SME were optimized using a grid search. Experimental results showed that the proposed SME effectively improved the weighing accuracy, and almost all of the egg weighing errors were less than 1 g with a processing speed of up to 5 eggs s-1. Compared with the AME, the overall mean error was reduced by approximately 86% to 93%, and the overall standard deviation of the error (SDE) was reduced by approximately 41% to 50%. Keywords: Asymmetrically trimmed mean, Digital filters, Dynamic weighing, Egg mass, Time-frequency analysis.

Terahertz (THz) technology has attracted great worldwide interest and novel high-intensity THz sources and plasmonics are two of the most active fields of recent research. Being situated between infrared light and microwave radiation, the absorption of THz rays in molecular and biomolecular systems is dominated by the excitation of intramolecular and intermolecular vibrations. This indicates that THz technology is an effective tool for sensing applications. However, the low sensitivity of free-space THz detection limits the sensing applications, which gives a great opportunity to metamaterials. Metamaterials are periodic artificial electromagnetic media structured with a size scale smaller than the wavelength of external stimuli. They present localized electric field enhancement and large values of quality factor (Q factor) and show high sensitivity to minor environment changes. In the present work, the mechanism of THz metamaterial sensing and dry sample and microfluidic sensing applications based on metamaterials are introduced. Moreover, new directions of THz metamaterial sensing advancement and introduction of two-dimensional materials and nanoparticles for future THz applications are summarized and discussed.

The development of chemometrics aims to provide an effective analysis approach for data generated by advanced analytical instruments. The success of existing analytical approaches in spectral analysis still relies on preprocessing and feature selection techniques to remove signal artifacts based on prior experiences. Data-driven deep learning analysis has been developed and successfully applied in many domains in the last few years. How to integrate deep learning with spectral analysis received increased attention for chemometrics. Approximately 20 recently published studies demonstrate that deep neural networks can learn critical patterns from raw spectra, which significantly reduces the demand for feature engineering. The composition of multiple processing layers improves the fitting and feature extraction capability and makes them applicable to various analytical tasks. This advance offers a new solution for chemometrics toward resolving challenges related to spectral data with rapidly increased sample numbers from various sources. We further provide a practical guide to the development of a deep convolutional neural network-based analytical workflow. The design of the network structure, tuning the hyperparameters in the training process, and repeatability of results is mainly discussed. Future studies are needed on interpretability and repeatability of the deep learning approach in spectral analysis.

The capability to design, fabricate, and optimize metamaterials based on various structures and material platforms has been crucial for the rapid development of modern terahertz (THz) technology. While the detailed structures of artificial unit cells within a metamaterial is certainly worth investigating, there has been increasing demand to integrate novel metamaterials with a traditional functional photonic device to form a hybrid device, whose performance is so significantly improved as to be promising for real-world applications. In this study, we proposed, for the first time, a THz parallel-plate resonator based on metallic mesh devices (MMDs) for chemical sensing applications. We studied the influences of various structural parameters through simulations, fabricated MMD-based resonator devices, and fully characterized the device performance through THz spectroscopy experiments. Furthermore, we experimentally demonstrated that our device can detect a doxycycline hydrochloride aqueous solution whose concentrations is as low as 1 mg L−1 through resonance frequency shifts, evidencing the device sensitivity capable of delicate chemical sensing tasks. Our work presents a practical and low cost architecture for chemical sensing using THz radiation, which opens new avenues for numerous useful THz devices based on metamaterials.

Impurity detection is crucial for quantitative image analysis in numerous food safety control and quality inspection applications. Rapid impurity detection of juglans using machine vision still faces challenge due to the complex shapes and color of foreign objects in different postures. Some traditional detection methods require expertly designed constraints and manual model parameters, and they have the poor detection performance and high model maintenance costs. In recent years, deep learning has become a focus in different research fields, because methods based on deep learning are able to directly learn features from training data. In this study, we originally proposed the two-stage convolutional networks to finish the image segmentation and detection of impurities in juglans images in real-time. The proposed segmentation method based on multiscale residual fully convolutional networks and classification method based on convolutional networks automatically can segment images and detect different-sized impurities (e.g., leaf debris, paper scraps, plastic scraps and metal parts) at the same time. The proposed deep-learning method is simpler and more effective, because it avoids extracting features manually, and it not only overcomes the conglomeration phenomenon between juglans and foreign objects in inline images, but also adapts to the disturb from surface abrasion damage on the white transmission belt to avoid error detection in the real factory environment. The proposed method is able to correctly segment 99.4% of the object regions in the test images and to correctly classify 96.5% of the foreign objects in the validation images and correctly detect 100.0% of test images. The segmentation and detection processing time of each image was less than 60 ms. Future work will focus on deep learning using multi-wave imaging and the sorting mechanical control.

Terahertz (THz) technique, with plenty attractive properties such as its prominent detecting ability, non-destruction and non-ionization, has been getting increased attention. THz modulator, enabling efficient manipulation, is crucial for THz devices and sensing applications. However, its practical implementation still remains a challenge because of the small modulation depth, slow modulation rate and difficult operation way. Nanomaterials have been constantly creating possibilities for the advancement of THz technology. Considering their excellent optical and electrical characteristics, gold nanorods (GNRs) are integrated with silicon-based materials to improve the performance for THz modulation, so as to promote the possibility of THz sensing in aqueous environment. In this paper, the modulation ability of THz wave based on bare silicon with/without dried GNR film, liquid cell with GNR solutions, and silicon channel with GNR solutions is systematically investigated. According to the results, a six-time larger modulation depth can be realized with the plasmonic effect of GNRs through infrared laser irradiation, which is employed to increase the protein detection sensitivity up to ppm level in water state. It is anticipated that integrating nanomaterials with THz functional devices is meaningful to improve the performance of THz technology, which could facilitate its wide applications in various fields.

Flexible plasmonic metasurfaces have garnered considerable attention because the material's mechanical flexibility enables new functionalities and integrated applications. Here, by adopting low-cost materials and simple techniques, we demonstrate a method of fabricating large flexible metasurfaces with arbitrary user-designed iridescent patterns. These naked-eye recognizable patterns together with their excellent plasmonic activities have yielded new functionalities and novel applications. Demonstrations include plasmonic sensing, reflective displays, developing new encryption strategies and integrated devices, etc. Moreover, the low fabrication cost (

Near-infrared (NIR) spectroscopy has been studied for over 100 years and has proven to be one of the most efficient technologies for determining the quality attributes of fruits. A number of reviews with regard to NIR technology and related applications have been reported. However, most of these reviews focus primarily on the basic principles, chemometrics, and applications, but nondestructive detection systems, especially for agricultural applications, are rarely discussed systematically and in detail. Therefore, this work provides a broad overview of NIR-based detection systems in terms of instrumentation (light source, monochromator, detector, sample compartment, and optics accessory), detection modes (reflectance, transmittance, and interactance), and system calibration procedures (wavelength range, spectral resolution, wavelength accuracy, stability, signal-to-noise ratio, and linearity). Furthermore, applications of NIR systems (laboratory systems, on-line systems, and portable systems) on fruits over the last decade are reviewed, as are the characteristics and limitations of these systems. Finally, some challenges in the use of NIR technology are discussed, and future work for NIR-based systems and applications is identified.

Quality evaluation of fruits and vegetables based on visible and near-infrared (Vis-NIR) spectroscopy has been investigated by many research groups worldwide during the last two decades. However, conventional Vis-NIR spectroscopy approximately measures the aggregate effect of absorption and scattering in fruits and vegetables without being able to separate them from each other. Optical property measurement, as an important means for analyzing or quantifying light scattering and propagation, would decouple and obtain absorption and scattering properties simultaneously. This article first gives a brief introduction to tissue optical properties, followed by a description of light transfer models. It also reviews recent developments of methods for measuring optical absorption and scattering properties of fruits and vegetables. Due to the complexity, diversity, and inhomogeneity of fruits and vegetables, major obstacles still exist for accurate measurement of the optical properties of fruits and vegetables composed of homogeneous or uniform layers. Furthermore, standard model samples with known optical properties, improved instrumentation, and better models for photon migration are needed for optical property measurement of fruits and vegetables. Research in the measurement and application of optical properties of fruits and vegetables is expected to grow in the future.