A Review of Deep Learning in Multiscale Agricultural Sensing

Reversible phase transformations between Pb nanocrystals and a viscous liquid-like phase

Phase transformations have been a prominent topic of study for both fundamental and applied science. Solid-liquid reaction-induced phase transformations can be hard to characterize, and the transformation mechanisms are often not fully understood. Here, we report reversible phase transformations between a metal (Pb) nanocrystal and a viscous liquid-like phase unveiled by in situ liquid cell transmission electron microscopy. The reversible phase transformations are obtained by modulating the electron current density (between 1000 and 3000 electrons Å-2 s-1). The metal-organic viscous liquid-like phase exhibits short-range ordering with a preferred Pb-Pb distance of 0.5 nm. Assisted by density functional theory and molecular dynamics calculations, we show that the viscous liquid-like phase results from the reactions of Pb with the CH3O fragments from the triethylene glycol solution under electron beam irradiation. Such reversible phase transformations may find broad implementations.

Joint superpixel and Transformer for high resolution remote sensing image classification

Deep neural networks combined with superpixel segmentation have proven to be superior to high-resolution remote sensing image (HRI) classification. Currently, most HRI classification methods that combine deep learning and superpixel segmentation use stacking on multiple scales to extract contextual information from segmented objects. However, this approach does not take into account the contextual dependencies between each segmented object. To solve this problem, a joint superpixel and Transformer (JST) framework is proposed for HRI classification. In JST, HRI is first segmented into superpixel objects as input, and Transformer is used to model the long-range dependencies. The contextual relationship between each input superpixel object is obtained and the class of analyzed objects is output by designing an encoding and decoding Transformer. Additionally, we explore the effect of semantic range on classification accuracy. JST is also tested by using two HRI datasets with overall classification accuracy, average accuracy and Kappa coefficients of 0.79, 0.70, 0.78 and 0.91, 0.85, 0.89, respectively. The effectiveness of the proposed method is compared qualitatively and quantitatively, and the results achieve competitive and consistently better than the benchmark comparison method.

Glyphosate: Impact on the microbiota-gut-brain axis and the immune-nervous system, and clinical cases of multiorgan toxicity

Glyphosate (GLP) and GLP-based herbicides (GBHs), such as polyethoxylated tallow amine-based GLP surfactants (GLP-SH), developed in the late 70', have become the most popular and controversial agrochemicals ever produced. Nowadays, GBHs have reached 350 million hectares of crops in over 140 countries, with an annual turnover of 5 billion and 11 billion USD in the U.S.A. and worldwide, respectively. Because of the highly efficient inhibitory activity of GLP targeted to the 5-enolpyruvylshikimate-3-phosphate synthase pathway, present in plants and several bacterial strains, the GLP-resistant crop-based genetic agricultural revolution has decreased famine and improved the costs and quality of living in developing countries. However, this progress has come at the cost of the 50-year GBH overuse, leading to environmental pollution, animal intoxication, bacterial resistance, and sustained occupational exposure of the herbicide farm and companies' workers. According to preclinical and clinical studies covered in the present review, poisoning with GLP, GLP-SH, and GBHs devastatingly affects gut microbiota and the microbiota-gut-brain (MGB) axis, leading to dysbiosis and gastrointestinal (GI) ailments, as well as immunosuppression and inappropriate immunostimulation, cholinergic neurotransmission dysregulation, neuroendocrinal system disarray, and neurodevelopmental and neurobehavioral alterations. Herein, we mainly focus on the contribution of gut microbiota (GM) to neurological impairments, e.g., stroke and neurodegenerative and neuropsychiatric disorders. The current review provides a comprehensive introduction to GLP's microbiological and neurochemical activities, including deviation of the intestinal Firmicutes-to-Bacteroidetes ratio, acetylcholinesterase inhibition, excitotoxicity, and mind-altering processes. Besides, it summarizes and critically discusses recent preclinical studies and clinical case reports concerning the harmful impacts of GBHs on the GI tract, MGB axis, and nervous system. Finally, an insightful comparison of toxic effects caused by GLP, GBH-SH, and GBHs is presented. To this end, we propose a first-to-date survey of clinical case reports on intoxications with these herbicides.

Deep learning framework for bovine iris segmentation

Iris segmentation is an initial step for identifying the biometrics of animals when establishing a traceability system for livestock. In this study, we propose a deep learning framework for pixel-wise segmentation of bovine iris with a minimized use of annotation labels utilizing the BovineAAEyes80 public dataset. The proposed image segmentation framework encompasses data collection, data preparation, data augmentation selection, training of 15 deep neural network (DNN) models with varying encoder backbones and segmentation decoder DNNs, and evaluation of the models using multiple metrics and graphical segmentation results. This framework aims to provide comprehensive and in-depth information on each model's training and testing outcomes to optimize bovine iris segmentation performance. In the experiment, U-Net with a VGG16 backbone was identified as the optimal combination of encoder and decoder models for the dataset, achieving an accuracy and dice coefficient score of 99.50% and 98.35%, respectively. Notably, the selected model accurately segmented even corrupted images without proper annotation data. This study contributes to the advancement of iris segmentation and the establishment of a reliable DNN training framework.

Multispectral Plant Disease Detection with Vision Transformer-Convolutional Neural Network Hybrid Approaches

Plant diseases pose a critical threat to global agricultural productivity, demanding timely detection for effective crop yield management. Traditional methods for disease identification are laborious and require specialised expertise. Leveraging cutting-edge deep learning algorithms, this study explores innovative approaches to plant disease identification, combining Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) to enhance accuracy. A multispectral dataset was meticulously collected to facilitate this research using six 50 mm filter filters, covering both the visible and several near-infrared (NIR) wavelengths. Among the models employed, ViT-B16 notably achieved the highest test accuracy, precision, recall, and F1 score across all filters, with averages of 83.3%, 90.1%, 90.75%, and 89.5%, respectively. Furthermore, a comparative analysis highlights the pivotal role of balanced datasets in selecting the appropriate wavelength and deep learning model for robust disease identification. These findings promise to advance crop disease management in real-world agricultural applications and contribute to global food security. The study underscores the significance of machine learning in transforming plant disease diagnostics and encourages further research in this field.

Chemometric Analysis for the Prediction of Biochemical Compounds in Leaves Using UV-VIS-NIR-SWIR Hyperspectroscopy

Reflectance hyperspectroscopy is recognised for its potential to elucidate biochemical changes, thereby enhancing the understanding of plant biochemistry. This study used the UV-VIS-NIR-SWIR spectral range to identify the different biochemical constituents in Hibiscus and Geranium plants. Hyperspectral vegetation indices (HVIs), principal component analysis (PCA), and correlation matrices provided in-depth insights into spectral differences. Through the application of advanced algorithms-such as PLS, VIP, iPLS-VIP, GA, RF, and CARS-the most responsive wavelengths were discerned. PLSR models consistently achieved R2 values above 0.75, presenting noteworthy predictions of 0.86 for DPPH and 0.89 for lignin. The red-edge and SWIR bands displayed strong associations with pivotal plant pigments and structural molecules, thus expanding the perspectives on leaf spectral dynamics. These findings highlight the efficacy of spectroscopy coupled with multivariate analysis in evaluating the management of biochemical compounds. A technique was introduced to measure the photosynthetic pigments and structural compounds via hyperspectroscopy across UV-VIS-NIR-SWIR, underpinned by rapid multivariate PLSR. Collectively, our results underscore the burgeoning potential of hyperspectroscopy in precision agriculture. This indicates a promising paradigm shift in plant phenotyping and biochemical evaluation.

Plant and Disease Recognition Based on PMF Pipeline Domain Adaptation Method: Using Bark Images as Meta-Dataset

Efficient image recognition is important in crop and forest management. However, it faces many challenges, such as the large number of plant species and diseases, the variability of plant appearance, and the scarcity of labeled data for training. To address this issue, we modified a SOTA Cross-Domain Few-shot Learning (CDFSL) method based on prototypical networks and attention mechanisms. We employed attention mechanisms to perform feature extraction and prototype generation by focusing on the most relevant parts of the images, then used prototypical networks to learn the prototype of each category and classify new instances. Finally, we demonstrated the effectiveness of the modified CDFSL method on several plant and disease recognition datasets. The results showed that the modified pipeline was able to recognize several cross-domain datasets using generic representations, and achieved up to 96.95% and 94.07% classification accuracy on datasets with the same and different domains, respectively. In addition, we visualized the experimental results, demonstrating the model's stable transfer capability between datasets and the model's high visual correlation with plant and disease biological characteristics. Moreover, by extending the classes of different semantics within the training dataset, our model can be generalized to other domains, which implies broad applicability.

Energy-Efficient IoT-Based Light Control System in Smart Indoor Agriculture

Indoor agriculture is emerging as a promising approach for increasing the efficiency and sustainability of agri-food production processes. It is currently evolving from a small-scale horticultural practice to a large-scale industry as a response to the increasing demand. This led to the appearance of plant factories where agri-food production is automated and continuous and the plant environment is fully controlled. While plant factories improve the productivity and sustainability of the process, they suffer from high energy consumption and the difficulty of providing the ideal environment for plants. As a small step to address these limitations, in this article we propose to use internet of things (IoT) technologies and automatic control algorithms to construct an energy-efficient remote control architecture for grow lights monitoring in indoor farming. The proposed architecture consists of using a master-slave device configuration in which the slave devices are used to control the local light conditions in growth chambers while the master device is used to monitor the plant factory through wireless communication with the slave devices. The devices all together make a 6LoWPAN network in which the RPL protocol is used to manage data transfer. This allows for the precise and centralized control of the growth conditions and the real-time monitoring of plants. The proposed control architecture can be associated with a decision support system to improve yields and quality at low costs. The developed method is evaluated in emulation software (Contiki-NG v4.7),its scalability to the case of large-scale production facilities is tested, and the obtained results are presented and discussed. The proposed approach is promising in dealing with control, cost, and scalability issues and can contribute to making smart indoor agriculture more effective and sustainable.

Non-Invasive Assessment, Classification, and Prediction of Biophysical Parameters Using Reflectance Hyperspectroscopy

Hyperspectral technology offers significant potential for non-invasive monitoring and prediction of morphological parameters in plants. In this study, UV-VIS-NIR-SWIR reflectance hyperspectral data were collected from Nicotiana tabacum L. plants using a spectroradiometer. These plants were grown under different light and gibberellic acid (GA₃) concentrations. Through spectroscopy and multivariate analyses, key growth parameters, such as height, leaf area, energy yield, and biomass, were effectively evaluated based on the interaction of light with leaf structures. The shortwave infrared (SWIR) bands, specifically SWIR1 and SWIR2, showed the strongest correlations with these growth parameters. When classifying tobacco plants grown under different GA₃ concentrations in greenhouses, artificial intelligence (AI) and machine learning (ML) algorithms were employed, achieving an average accuracy of over 99.1% using neural network (NN) and gradient boosting (GB) algorithms. Among the 34 tested vegetation indices, the photochemical reflectance index (PRI) demonstrated the strongest correlations with all evaluated plant phenotypes. Partial least squares regression (PLSR) models effectively predicted morphological attributes, with R²_CV values ranging from 0.81 to 0.87 and RPD_P values exceeding 2.09 for all parameters. Based on Pearson's coefficient XYZ interpolations and HVI algorithms, the NIR-SWIR band combination proved the most effective for predicting height and leaf area, while VIS-NIR was optimal for optimal energy yield, and VIS-VIS was best for predicting biomass. To further corroborate these findings, the SWIR bands for certain morphological characteristic wavelengths selected with s-PLS were most significant for SWIR1 and SWIR2, while i-PLS showed a more uniform distribution in VIS-NIR-SWIR bands. Therefore, SWIR hyperspectral bands provide valuable insights into developing alternative bands for remote sensing measurements to estimate plant morphological parameters. These findings underscore the potential of remote sensing technology for rapid, accurate, and non-invasive monitoring within stationary high-throughput phenotyping systems in greenhouses. These insights align with advancements in digital and precision technology, indicating a promising future for research and innovation in this field.

Glyphosate Separating and Sensing for Precision Agriculture and Environmental Protection in the Era of Smart Materials

The present article critically and comprehensively reviews the most recent reports on smart sensors for determining glyphosate (GLP), an active agent of GLP-based herbicides (GBHs) traditionally used in agriculture over the past decades. Commercialized in 1974, GBHs have now reached 350 million hectares of crops in over 140 countries with an annual turnover of 11 billion USD worldwide. However, rolling exploitation of GLP and GBHs in the last decades has led to environmental pollution, animal intoxication, bacterial resistance, and sustained occupational exposure of the herbicide of farm and companies' workers. Intoxication with these herbicides dysregulates the microbiome-gut-brain axis, cholinergic neurotransmission, and endocrine system, causing paralytic ileus, hyperkalemia, oliguria, pulmonary edema, and cardiogenic shock. Precision agriculture, i.e., an (information technology)-enhanced approach to crop management, including a site-specific determination of agrochemicals, derives from the benefits of smart materials (SMs), data science, and nanosensors. Those typically feature fluorescent molecularly imprinted polymers or immunochemical aptamer artificial receptors integrated with electrochemical transducers. Fabricated as portable or wearable lab-on-chips, smartphones, and soft robotics and connected with SM-based devices that provide machine learning algorithms and online databases, they integrate, process, analyze, and interpret massive amounts of spatiotemporal data in a user-friendly and decision-making manner. Exploited for the ultrasensitive determination of toxins, including GLP, they will become practical tools in farmlands and point-of-care testing. Expectedly, smart sensors can be used for personalized diagnostics, real-time water, food, soil, and air quality monitoring, site-specific herbicide management, and crop control.

Reflectance Spectroscopy for the Classification and Prediction of Pigments in Agronomic Crops

Reflectance spectroscopy, in combination with machine learning and artificial intelligence algorithms, is an effective method for classifying and predicting pigments and phenotyping in agronomic crops. This study aims to use hyperspectral data to develop a robust and precise method for the simultaneous evaluation of pigments, such as chlorophylls, carotenoids, anthocyanins, and flavonoids, in six agronomic crops: corn, sugarcane, coffee, canola, wheat, and tobacco. Our results demonstrate high classification accuracy and precision, with principal component analyses (PCAs)-linked clustering and a kappa coefficient analysis yielding results ranging from 92 to 100% in the ultraviolet-visible (UV-VIS) to near-infrared (NIR) to shortwave infrared (SWIR) bands. Predictive models based on partial least squares regression (PLSR) achieved R² values ranging from 0.77 to 0.89 and ratio of performance to deviation (RPD) values over 2.1 for each pigment in C₃ and C₄ plants. The integration of pigment phenotyping methods with fifteen vegetation indices further improved accuracy, achieving values ranging from 60 to 100% across different full or range wavelength bands. The most responsive wavelengths were selected based on a cluster heatmap, β-loadings, weighted coefficients, and hyperspectral vegetation index (HVI) algorithms, thereby reinforcing the effectiveness of the generated models. Consequently, hyperspectral reflectance can serve as a rapid, precise, and accurate tool for evaluating agronomic crops, offering a promising alternative for monitoring and classification in integrated farming systems and traditional field production. It provides a non-destructive technique for the simultaneous evaluation of pigments in the most important agronomic plants.

Multi-Omics Pipeline and Omics-Integration Approach to Decipher Plant's Abiotic Stress Tolerance Responses

The present day's ongoing global warming and climate change adversely affect plants through imposing environmental (abiotic) stresses and disease pressure. The major abiotic factors such as drought, heat, cold, salinity, etc., hamper a plant's innate growth and development, resulting in reduced yield and quality, with the possibility of undesired traits. In the 21st century, the advent of high-throughput sequencing tools, state-of-the-art biotechnological techniques and bioinformatic analyzing pipelines led to the easy characterization of plant traits for abiotic stress response and tolerance mechanisms by applying the 'omics' toolbox. Panomics pipeline including genomics, transcriptomics, proteomics, metabolomics, epigenomics, proteogenomics, interactomics, ionomics, phenomics, etc., have become very handy nowadays. This is important to produce climate-smart future crops with a proper understanding of the molecular mechanisms of abiotic stress responses by the plant's genes, transcripts, proteins, epigenome, cellular metabolic circuits and resultant phenotype. Instead of mono-omics, two or more (hence 'multi-omics') integrated-omics approaches can decipher the plant's abiotic stress tolerance response very well. Multi-omics-characterized plants can be used as potent genetic resources to incorporate into the future breeding program. For the practical utility of crop improvement, multi-omics approaches for particular abiotic stress tolerance can be combined with genome-assisted breeding (GAB) by being pyramided with improved crop yield, food quality and associated agronomic traits and can open a new era of omics-assisted breeding. Thus, multi-omics pipelines together are able to decipher molecular processes, biomarkers, targets for genetic engineering, regulatory networks and precision agriculture solutions for a crop's variable abiotic stress tolerance to ensure food security under changing environmental circumstances.

A Systematic Solution for Moving-Target Detection and Tracking While Only Using a Monocular Camera

This paper focuses on moving-target detection and tracking in a three-dimensional (3D) space, and proposes a visual target tracking system only using a two-dimensional (2D) camera. To quickly detect moving targets, an improved optical flow method with detailed modifications in the pyramid, warping, and cost volume network (PWC-Net) is applied. Meanwhile, a clustering algorithm is used to accurately extract the moving target from a noisy background. Then, the target position is estimated using a proposed geometrical pinhole imaging algorithm and cubature Kalman filter (CKF). Specifically, the camera's installation position and inner parameters are applied to calculate the azimuth, elevation angles, and depth of the target while only using 2D measurements. The proposed geometrical solution has a simple structure and fast computational speed. Different simulations and experiments verify the effectiveness of the proposed method.

Biophysical, Biochemical, and Photochemical Analyses Using Reflectance Hyperspectroscopy and Chlorophyll a Fluorescence Kinetics in Variegated Leaves

The adjustments that occur during photosynthesis are correlated with morphological, biochemical, and photochemical changes during leaf development. Therefore, monitoring leaves, especially when pigment accumulation occurs, is crucial for monitoring organelles, cells, tissue, and whole-plant levels. However, accurately measuring these changes can be challenging. Thus, this study tests three hypotheses, whereby reflectance hyperspectroscopy and chlorophyll a fluorescence kinetics analyses can improve our understanding of the photosynthetic process in Codiaeum variegatum (L.) A. Juss, a plant with variegated leaves and different pigments. The analyses include morphological and pigment profiling, hyperspectral data, chlorophyll a fluorescence curves, and multivariate analyses using 23 JIP test parameters and 34 different vegetation indexes. The results show that photochemical reflectance index (PRI) is a useful vegetation index (VI) for monitoring biochemical and photochemical changes in leaves, as it strongly correlates with chlorophyll and nonphotochemical dissipation (Kn) parameters in chloroplasts. In addition, some vegetation indexes, such as the pigment-specific simple ratio (PSSRc), anthocyanin reflectance index (ARI1), ratio analysis of reflectance spectra (RARS), and structurally insensitive pigment index (SIPI), are highly correlated with morphological parameters and pigment levels, while PRI, moisture stress index (MSI), normalized difference photosynthetic (PVR), fluorescence ratio (FR), and normalized difference vegetation index (NDVI) are associated with photochemical components of photosynthesis. Combined with the JIP test analysis, our results showed that decreased damage to energy transfer in the electron transport chain is correlated with the accumulation of carotenoids, anthocyanins, flavonoids, and phenolic compounds in the leaves. Phenomenological energy flux modelling shows the highest changes in the photosynthetic apparatus based on PRI and SIPI when analyzed with Pearson's correlation, the hyperspectral vegetation index (HVI) algorithm, and the partial least squares (PLS) to select the most responsive wavelengths. These findings are significant for monitoring nonuniform leaves, particularly when leaves display high variation in pigment profiling in variegated and colorful leaves. This is the first study on the rapid and precise detection of morphological, biochemical, and photochemical changes combined with vegetation indexes for different optical spectroscopy techniques.

Meta-learning shows great potential in plant disease recognition under few available samples

Plant diseases worsen the threat of food shortage with the growing global population, and disease recognition is the basis for the effective prevention and control of plant diseases. Deep learning has made significant breakthroughs in the field of plant disease recognition. Compared with traditional deep learning, meta-learning can still maintain more than 90% accuracy in disease recognition with small samples. However, there is no comprehensive review on the application of meta-learning in plant disease recognition. Here, we mainly summarize the functions, advantages, and limitations of meta-learning research methods and their applications for plant disease recognition with a few data scenarios. Finally, we outline several research avenues for utilizing current and future meta-learning in plant science. This review may help plant science researchers obtain faster, more accurate, and more credible solutions through deep learning with fewer labeled samples.

Enhancing Pigment Phenotyping and Classification in Lettuce through the Integration of Reflectance Spectroscopy and AI Algorithms

In this study, we investigated the use of artificial intelligence algorithms (AIAs) in combination with VIS-NIR-SWIR hyperspectroscopy for the classification of eleven lettuce plant varieties. For this purpose, a spectroradiometer was utilized to collect hyperspectral data in the VIS-NIR-SWIR range, and 17 AIAs were applied to classify lettuce plants. The results showed that the highest accuracy and precision were achieved using the full hyperspectral curves or the specific spectral ranges of 400-700 nm, 700-1300 nm, and 1300-2400 nm. Four models, AdB, CN2, G-Boo, and NN, demonstrated exceptional R² and ROC values, exceeding 0.99, when compared between all models and confirming the hypothesis and highlighting the potential of AIAs and hyperspectral fingerprints for efficient, precise classification and pigment phenotyping in agriculture. The findings of this study have important implications for the development of efficient methods for phenotyping and classification in agriculture and the potential of AIAs in combination with hyperspectral technology. To advance our understanding of the capabilities of hyperspectroscopy and AIs in precision agriculture and contribute to the development of more effective and sustainable agriculture practices, further research is needed to explore the full potential of these technologies in different crop species and environments.

Assessment of Black Rot in Oilseed Rape Grown under Climate Change Conditions Using Biochemical Methods and Computer Vision

Global warming is a challenge for plants and pathogens, involving profound changes in the physiology of both contenders to adapt to the new environmental conditions and to succeed in their interaction. Studies have been conducted on the behavior of oilseed rape plants and two races (1 and 4) of the bacterium Xanthomonas campestris pv. campestris (Xcc) and their interaction to anticipate our response in the possible future climate. Symptoms caused by both races of Xcc were very similar to each other under any climatic condition assayed, although the bacterial count from infected leaves differed for each race. Climate change caused an earlier onset of Xcc symptoms by at least 3 days, linked to oxidative stress and a change in pigment composition. Xcc infection aggravated the leaf senescence already induced by climate change. To identify Xcc-infected plants early under any climatic condition, four classifying algorithms were trained with parameters obtained from the images of green fluorescence, two vegetation indices and thermography recorded on Xcc-symptomless leaves. Classification accuracies were above 0.85 out of 1.0 in all cases, with k-nearest neighbor analysis and support vector machines performing best under the tested climatic conditions.

Plant growth-promoting rhizobacteria: Salt stress alleviators to improve crop productivity for sustainable agriculture development

Soil salinity, a growing issue worldwide, is a detrimental consequence of the ever-changing climate, which has highlighted and worsened the conditions associated with damaged soil quality, reduced agricultural production, and decreasing land areas, thus resulting in an unsteady national economy. In this review, halo-tolerant plant growth-promoting rhizo-microbiomes (PGPRs) are evaluated in the salinity-affected agriculture as they serve as excellent agents in controlling various biotic-abiotic stresses and help in the augmentation of crop productivity. Integrated efforts of these effective microbes lighten the load of agro-chemicals on the environment while managing nutrient availability. PGPR-assisted modern agriculture practices have emerged as a green strategy to benefit sustainable farming without compromising the crop yield under salinity as well as salinity-affected supplementary stresses including increased temperature, drought, salinity, and potential invasive plant pathogenicity. PGPRs as bio-inoculants impart induced systemic tolerance (IST) to plants by the production of volatile organic compounds (VOCs), antioxidants, osmolytes, extracellular polymeric substances (EPS), phytohormones, and ACC-deaminase and recuperation of nutritional status and ionic homeostasis. Regulation of PGPR-induced signaling pathways such as MAPK and CDPK assists in salinity stress alleviation. The "Next Gen Agriculture" consists of the application of designer crop microbiomes through gene editing tools, for instance, CRISPR, and engineering of the metabolic pathways of the microbes so as to gain maximum plant resistance. The utilization of omics technologies over the traditional approaches can fulfill the criteria required to increase crop yields in a sustainable manner for feeding the burgeoning population and augment plant adaptability under climate change conditions, ultimately leading to improved vitality. Furthermore, constraints such as the crop specificity issue of PGPR, lack of acceptance by farmers, and legal regulatory aspects have been acknowledged while also discussing the future trends for product commercialization with the view of the changing climate.

Designing a Proximal Sensing Camera Acquisition System for Vineyard Applications: Results and Feedback on 8 Years of Experiments

The potential of image proximal sensing for agricultural applications has been a prolific scientific subject in the recent literature. Its main appeal lies in the sensing of precise information about plant status, which is either harder or impossible to extract from lower-resolution downward-looking image sensors such as satellite or drone imagery. Yet, many theoretical and practical problems arise when dealing with proximal sensing, especially on perennial crops such as vineyards. Indeed, vineyards exhibit challenging physical obstacles and many degrees of variability in their layout. In this paper, we present the design of a mobile camera suited to vineyards and harsh experimental conditions, as well as the results and assessments of 8 years' worth of studies using that camera. These projects ranged from in-field yield estimation (berry counting) to disease detection, providing new insights on typical viticulture problems that could also be generalized to orchard crops. Different recommendations are then provided using small case studies, such as the difficulties related to framing plots with different structures or the mounting of the sensor on a moving vehicle. While results stress the obvious importance and strong benefits of a thorough experimental design, they also indicate some inescapable pitfalls, illustrating the need for more robust image analysis algorithms and better databases. We believe sharing that experience with the scientific community can only benefit the future development of these innovative approaches.

A Review of Hybrid Approaches for Quantitative Assessment of Crop Traits Using Optical Remote Sensing: Research Trends and Future Directions

Remote sensing technology allows to provide information about biochemical and biophysical crop traits and monitor their spatiotemporal dynamics of agriculture ecosystems. Among multiple retrieval techniques, hybrid approaches have been found to provide outstanding accuracy, for instance, for the inference of leaf area index (LAI), fractional vegetation cover (fCover), and leaf and canopy chlorophyll content (LCC and CCC). The combination of radiative transfer models (RTMs) and data-driven models creates an advantage in the use of hybrid methods. Through this review paper, we aim to provide state-of-the-art hybrid retrieval schemes and theoretical frameworks. To achieve this, we reviewed and systematically analyzed publications over the past 22 years. We identified two hybrid-based parametric and hybrid-based nonparametric regression models and evaluated their performance for each variable of interest. From the results of our extensive literature survey, most research directions are now moving towards combining RTM and machine learning (ML) methods in a symbiotic manner. In particular, the development of ML will open up new ways to integrate innovative approaches such as integrating shallow or deep neural networks with RTM using remote sensing data to reduce errors in crop trait estimations and improve control of crop growth conditions in very large areas serving precision agriculture applications.

Machine Learning in the Analysis of Multispectral Reads in Maize Canopies Responding to Increased Temperatures and Water Deficit

Real-time monitoring of crop responses to environmental deviations represents a new avenue for applications of remote and proximal sensing. Combining the high-throughput devices with novel machine learning (ML) approaches shows promise in the monitoring of agricultural production. The 3 × 2 multispectral arrays with responses at 610 and 680 nm (red), 730 and 760 nm (red-edge) and 810 and 860 nm (infrared) spectra were used to assess the occurrence of leaf rolling (LR) in 545 experimental maize plots measured four times for calibration dataset (n = 2180) and 145 plots measured once for external validation. Multispectral reads were used to calculate 15 simple normalized vegetation indices. Four ML algorithms were assessed: single and multilayer perceptron (SLP and MLP), convolutional neural network (CNN) and support vector machines (SVM) in three validation procedures, which were stratified cross-validation, random subset validation and validation with external dataset. Leaf rolling occurrence caused visible changes in spectral responses and calculated vegetation indexes. All algorithms showed good performance metrics in stratified cross-validation (accuracy >80%). SLP was the least efficient in predictions with external datasets, while MLP, CNN and SVM showed comparable performance. Combining ML with multispectral sensing shows promise in transition towards agriculture based on data-driven decisions especially considering the novel Internet of Things (IoT) avenues.

Investigating the Potential of Sentinel-2 MSI in Early Crop Identification in Northeast China

Early crop identification can provide timely and valuable information for agricultural planting management departments to make reasonable and correct decisions. At present, there is still a lack of systematic summary and analysis on how to obtain real-time samples in the early stage, what the optimal feature sets are, and what level of crop identification accuracy can be achieved at different stages. First, this study generated training samples with the help of historical crop maps in 2019 and remote sensing images in 2020. Then, a feature optimization method was used to obtain the optimal features in different stages. Finally, the differences of the four classifiers in identifying crops and the variation characteristics of crop identification accuracy at different stages were analyzed. These experiments were conducted at three sites in Heilongjiang Province to evaluate the reliability of the results. The results showed that the earliest identification time of corn can be obtained in early July (the seven leaves period) with an identification accuracy up to 86%. In the early stages, its accuracy was 40~79%, which was low, and could not reach the satisfied accuracy requirements. In the middle stages, a satisfactory recognition accuracy could be achieved, and its recognition accuracy was 79~100%. The late stage had a higher recognition accuracy, which was 90~100%. The accuracy of soybeans at each stage was similar to that of corn, and the earliest identification time of soybeans could also be obtained in early July (the blooming period) with an identification accuracy up to 87%. Its accuracy in the early growth stage was 35~71%; in the middle stage, it was 69~100%; and in the late stage, it was 92~100%. Unlike corn and soybeans, the earliest identification time of rice could be obtained at the end of April (the flooding period) with an identification accuracy up to 86%. In the early stage, its accuracy was 58~100%; in the middle stage, its accuracy was 93~100%; and in the late stage, its accuracy was 96~100%. In terms of crop identification accuracy in the whole growth stage, GBDT and RF performed better than other classifiers in our three study areas. This study systematically investigated the potential of early crop recognition in Northeast China, and the results are helpful for relevant applications and decision making of crop recognition in different crop growth stages.

Digital Transformation in Smart Farm and Forest Operations Needs Human-Centered AI: Challenges and Future Directions

The main impetus for the global efforts toward the current digital transformation in almost all areas of our daily lives is due to the great successes of artificial intelligence (AI), and in particular, the workhorse of AI, statistical machine learning (ML). The intelligent analysis, modeling, and management of agricultural and forest ecosystems, and of the use and protection of soils, already play important roles in securing our planet for future generations and will become irreplaceable in the future. Technical solutions must encompass the entire agricultural and forestry value chain. The process of digital transformation is supported by cyber-physical systems enabled by advances in ML, the availability of big data and increasing computing power. For certain tasks, algorithms today achieve performances that exceed human levels. The challenge is to use multimodal information fusion, i.e., to integrate data from different sources (sensor data, images, *omics), and explain to an expert why a certain result was achieved. However, ML models often react to even small changes, and disturbances can have dramatic effects on their results. Therefore, the use of AI in areas that matter to human life (agriculture, forestry, climate, health, etc.) has led to an increased need for trustworthy AI with two main components: explainability and robustness. One step toward making AI more robust is to leverage expert knowledge. For example, a farmer/forester in the loop can often bring in experience and conceptual understanding to the AI pipeline—no AI can do this. Consequently, human-centered AI (HCAI) is a combination of “artificial intelligence” and “natural intelligence” to empower, amplify, and augment human performance, rather than replace people. To achieve practical success of HCAI in agriculture and forestry, this article identifies three important frontier research areas: (1) intelligent information fusion; (2) robotics and embodied intelligence; and (3) augmentation, explanation, and verification for trusted decision support. This goal will also require an agile, human-centered design approach for three generations (G). G1: Enabling easily realizable applications through immediate deployment of existing technology. G2: Medium-term modification of existing technology. G3: Advanced adaptation and evolution beyond state-of-the-art.

An AUV Target-Tracking Method Combining Imitation Learning and Deep Reinforcement Learning

This study aims to solve the problem of sparse reward and local convergence when using a reinforcement learning algorithm as the controller of an AUV. Based on the generative adversarial imitation (GAIL) algorithm combined with a multi-agent, a multi-agent GAIL (MAG) algorithm is proposed. The GAIL enables the AUV to directly learn from expert demonstrations, overcoming the difficulty of slow initial training of the network. Parallel training of multi-agents reduces the high correlation between samples to avoid local convergence. In addition, a reward function is designed to help training. Finally, the results show that in the unity simulation platform test, the proposed algorithm has a strong optimal decision-making ability in the tracking process.