Fine hyperspectral classification of rice varieties based on self-attention mechanism

The accurate identification of rice varieties using rapid and nondestructive hyperspectral technology is of practical significance for rice cultivation and agricultural production. This paper proposes a convolutional neural network classification model based on a self-attention mechanism (self-attention-1D-CNN) to improve accuracy in distinguishing between crop species in fields using canopy spectral information. After experimental materials were planted in the research area, portable equipment was used to collect the canopy hyperspectral data for rice during the booting stage. Five preprocessing methods and three extraction methods were used to process the data. A comparison of the classification accuracy of different classification models showed that the self-attention-1D-CNN proposed in this study achieved the best classification with an accuracy of 99.93%. The research demonstrated the feasibility of using hyperspectral technology for the fine classification of rice varieties, and the feasibility of using the CNN model as a potential classification method for near-ground crop monitoring and classification.     

Soil suitability classification for crop selection in precision agriculture using GBRT-based hybrid DNN surrogate models

The main reason for agricultural productivity decline is farmers' failure to choose the appropriate crop for their soil. It is important for farmers to understand which crops are suitable for different soil types based on their characteristics. Due to the vast variety of soil types worldwide, farmers often struggle to choose the most profitable crop for their land. To improve crop yields, a crop selection system has been developed using GBRT-based deep learning surrogate models. Gradient Boosted Regression Tree (GBRT) has been combined with a Bayesian optimization (BO) algorithm to determine the most optimal hyperparameters for the deep neural network. The optimized hyperparameters are then applied during the testing phase. Further, the impact of each input parameter on the individual outputs is evaluated using explainable artificial intelligence (XAI). The crop recommendation system comprises data preparation, classification, and performance evaluation modules. A classification method based on confusion matrices and performance matrices, as well as feature analysis using density plots and correlation plots, follows. The crop selection system categorizes the experimental dataset into 12 classes, with three for each of the four crops. The dataset includes soil-specific physical and chemical features such as sand, silt, clay, pH, electric conductivity (EC), soil organic carbon (SOC), nitrogen (N), phosphorus (P), and potassium (K). The developed surrogate model is highly accurate, precise, and reliable, with an F1-Score of 1.0 for all classes in the dataset, indicating exact accuracy and recall. The DNN-based classification model achieves an average classification accuracy of 1.00.     

Fast Evaluation of Unhealthy and Healthy Neonates Using Hyperspectral Features on 700-850 Nm Wavelengths,ROI Extraction,and 3D-CNN

ObjectivesHyperspectral imaging (HSI) has great potential in detecting the health conditions of neonates as it provides diagnostic information about the tissue by avoiding tissue biopsy. HSI gives more features than thermal imaging, which can obtain images in a single wavelength, as it can obtain images in a large number of wavelengths. The data obtained with hyperspectral sensors are 3-dimensional data called hypercube including first two-dimensional spatial information and third-dimensional spectral information.Material and methodsIn this study, hyperspectral data were obtained from 19 different neonates in the Neonatal Intensive Care Unit (NICU) of Selcuk University, Medical Faculty. There are 16 hypercubes from 16 unhealthy neonates, 16 hypercubes from 3 healthy neonates in a period of three months, and 32 hypercubes in total are available. For the training of 3D-CNN model, data augmentation methods, such as rotation, height shifting, width shifting, and shearing were applied to hyperspectral data. A number of 32 hypercubes taken from neonates in NICU were augmented to 160 hypercubes. Spectral signatures were examined and 51 bands in the range of 700-850 nm with distinctive features were used for the classification. The spectral dimension was reduced by applying Principal Component Analysis (PCA) to all hypercubes. In addition, it is aimed to obtain both spectral and spatial features with the 3D-CNN. For increasing the classification efficiency, ROI extraction was made and four datasets were created in different spatial dimensions. These datasets contain 160, 640, 1440, and 5760 hypercubes, respectively.ResultsThe best result was achieved by using 5760 hypercubes of 25x25x51. As a result of the classification of the hypercubes, accuracy 98.00%, sensitivity 97.22%, and specificity 98.78% were obtained. It was determined how many PCs used to achieve the best result. Further, the proposed 3D-CNN model is compared to 2D-CNN model to evaluate the performance of the study.ConclusionIt was aimed to evaluate the health status of neonates fastly by using HSI and 3D-CNN for the first time. The obtained results are an indication that HSI and 3D-CNN are very effective for the evaluation of unhealthy and healthy neonates.     

沙地油蒿群落覆盖度的遥感定量化研究

鄂尔多斯地区荒漠化与沙地植被覆盖度有紧密的关系。用 NOAA卫星数据对沙地植被覆盖度进行动态监测可以为认识该地区的荒漠化动态变化及过程提供丰富的信息。通过分析沙地反射机理 ,建立了沙地油蒿(Artemisia ordosica)群落盖度与修正后的土壤调节植被指数 (MSAVI)之间的关系模型。然而 ,卫星数据用于该关系模型之前 ,必须进行处理。本文首先用裸沙土壤线法校正大气影响 ,然后 ,用 MSAVI消除土壤背景影响     

Yield responses of crops to changes in environment and management practices: Model sensitivity analysis. I. Maize

YIELD, a parametric crop production model, employs climatic data to calculate actual and potential yield for various crops and includes formulations for specific crop and growth stage effects. The objective was to demonstrate the sensitivity of YIELD for grain corn (maize) to changes in various environmental and decision-making inputs. Five temperature, five solar radiation, six relative humidity regimes, five water application schemes, and four irrigation frequencies were included in this study. The effects of different soil types and wind regimes on crop water requirements were investigated. The model output includes crop yield, water use efficiency, and management efficiency. Among the results, yield decreased on the average by 3.9% per one degree (C) increase in air temperature. A 1% change in solar radiation resulted in an average of 1% change in yield. Similar changes in relative humidity caused a yield change of about 0.8%.     

Hierarchical Bayes estimation of hunting success rates with spatial correlations

A Bayesian hierarchical generalized linear model is used to estimate hunting success rates at the subarea level for postseason harvest surveys. The model includes fixed week effects, random geographic effects, and spatial correlations between neighboring subareas. The computation is done by Gibbs sampling and adaptive rejection sampling techniques. The method is illustrated using data from the Missouri Turkey Hunting Survey in the spring of 1996. Bayesian model selection methods are used to demonstrate that there are significant week differences and spatial correlations of hunting success rates among counties. The Bayesian estimates are also shown to be quite robust in terms of changes of hyperparameters.     

Parameter optimization and field validation of the functional-structural model GREENLAB for maize

BACKGROUND AND AIMS: There are three reasons for the increasing demand for crop models that build the plant on the basis of architectural principles and organogenetic processes: (1) realistic concepts for developing new crops need to be guided by such models; (2) there is an increasing interest in crop phenotypic plasticity, based on variable architecture and morphology; and (3) engineering of mechanized cropping systems requires information on crop architecture. The functional-structural model GREENLAB was recently presented that simulates resource-dependent plasticity of plant architecture. This study introduces a new methodology for crop parameter optimization against measured data called multi-fitting, validates the calibrated model for maize with independent field data, and describes a technique for 3D visualization of outputs. METHODS: Maize was grown near Beijing during the 2000, 2001 and 2003 (two sowing dates) summer seasons in a block design with four to five replications. Detailed morphological and topological observations were made on the plant architecture throughout the development of the four crops. Data obtained in 2000 was used to establish target files for parameter optimization using the generalized least square method, and parameter accuracy was evaluated by coefficient of variance. In situ plant digitization was used to establish 3D symbol files for organs that were then used to translate model outputs directly into 3D representations for each time step of model execution. KEY RESULTS AND CONCLUSIONS: Multi-fitting against several target files obtained at different growth stages gave better parameter accuracy than single fitting at maturity only, and permitted extracting generic organ expansion kinetics from the static observations. The 2000 model gave excellent predictions of plant architecture and vegetative growth for the other three seasons having different temperature regimes, but predictions of inter-seasonal variability of biomass partitioning during grain filling were less accurate. This was probably due to insufficient consideration of processes governing cob sink size and terminal leaf senescence. Further perspectives for model improvement are discussed.     

农作物种质资源登记区块链模型研究

农作物种质资源工作发展至今取得了巨大的成就,但也存在着农作物种质资源家底不清、共享利用效率不高及种质权属不清等问题,开展农作物种质资源登记工作能够有效解决以上问题。本研究从资源登记信息的流向角度出发,梳理了登记的整体工作流程,并结合区块链技术的优势,提出了农作物种质资源登记区块链网络模型。首先根据农作物种质资源登记的业务特点,确定登记主体和登记流程,从数据的填报、审核与共享3个阶段阐明了资源登记的整体流程。在此基础上设计了农作物种质资源登记区块链网络模型、工作流模型和数据模型,并根据农作物种质资源实际工作需求改善了DPOS共识机制,给出了激励机制的构建方式。农作物种质资源登记区块链模型通过去中心化实现登记数据的分布式存储,通过共识机制增加数据的可靠性,通过区块链的特殊数据结构确保链中数据难篡改、易追溯,提高数据安全性,为今后资源登记系统的设计提供了新思路。     

Simulating Crop Evapotranspiration Response under Different Planting Scenarios by Modified SWAT Model in an Irrigation District,Northwest China

Modelling crop evapotranspiration (ET) response to different planting scenarios in an irrigation district plays a significant role in optimizing crop planting patterns, resolving agricultural water scarcity and facilitating the sustainable use of water resources. In this study, the SWAT model was improved by transforming the evapotranspiration module. Then, the improved model was applied in Qingyuan Irrigation District of northwest China as a case study. Land use, soil, meteorology, irrigation scheduling and crop coefficient were considered as input data, and the irrigation district was divided into subdivisions based on the DEM and local canal systems. On the basis of model calibration and verification, the improved model showed better simulation efficiency than did the original model. Therefore, the improved model was used to simulate the crop evapotranspiration response under different planting scenarios in the irrigation district. Results indicated that crop evapotranspiration decreased by 2.94% and 6.01% under the scenarios of reducing the planting proportion of spring wheat (scenario 1) and summer maize (scenario 2) by keeping the total cultivated area unchanged. However, the total net output values presented an opposite trend under different scenarios. The values decreased by 3.28% under scenario 1, while it increased by 7.79% under scenario 2, compared with the current situation. This study presents a novel method to estimate crop evapotranspiration response under different planting scenarios using the SWAT model, and makes recommendations for strategic agricultural water management planning for the rational utilization of water resources and development of local economy by studying the impact of planting scenario changes on crop evapotranspiration and output values in the irrigation district of northwest China.     

NeEstimator v2: re‐implementation of software for the estimation of contemporary effective population size (Ne) from genetic data

NeEstimator v2 is a completely revised and updated implementation of software that produces estimates of contemporary effective population size, using several different methods and a single input file. NeEstimator v2 includes three single‐sample estimators (updated versions of the linkage disequilibrium and heterozygote‐excess methods, and a new method based on molecular coancestry), as well as the two‐sample (moment‐based temporal) method. New features include the following: (i) an improved method for accounting for missing data; (ii) options for screening out rare alleles; (iii) confidence intervals for all methods; (iv) the ability to analyse data sets with large numbers of genetic markers (10 000 or more); (v) options for batch processing large numbers of different data sets, which will facilitate cross‐method comparisons using simulated data; and (vi) correction for temporal estimates when individuals sampled are not removed from the population (Plan I sampling). The user is given considerable control over input data and composition, and format of output files. The freely available software has a new JAVA interface and runs under MacOS, Linux and Windows.     

利用遥感技术实现作物模拟模型区域应用的研究进展

作物模拟模型从单点发展到区域应用时,模型中一些宏观资料的获取和参数的区域化方面出现困难,利用遥感技术将实现作物模拟模型的区域应用．文中综述了近年来遥感反演作物模型所需的地表生物物理参数的方法、利用遥感信息直接获取生物量的途径和遥感信息与作物模拟模型之间时空匹配问题等方面的研究概况,重点介绍了利用遥感技术实现作物模拟模型区域应用的3种解决方案(强迫型、调控型和验证型)及其研究进展,并讨论了目前存在的问题和今后研究的方向．     

Monitoring rapid vegetation succession in estuarine wetland using time series MODIS-based indicators: An application in the Yangtze River Delta area

Frequent and continuous time series is required for the detection of plant phenology and vegetation succession. The launch of novel remote sensor MODIS (moderate resolution imaging spectroradiometer) provided us with an opportunity to make a new trial of studying the rapid vegetation succession in estuarine wetlands. In this study, the spatiotemporal variations of vegetation cover and tidal flat elevation along a transect (covering 6 pixels of MODIS) of an estuarine wetland at Dongtan, Chongming Island, in Yangtze River estuary, China were investigated to assess its rapid vegetation succession and physical conditions. By combining the field data collected, the time series of MODIS-based VIs (vegetation indices), including NDVI (normalized difference vegetation index), EVI (enhanced vegetation index) and MSAVI (modified soil adjusted vegetation index), and a water index, LSWI (land surface water index) were utilized to characterize the rapid vegetation succession between 2001 and 2006. We found that NDVI, EVI and MSAVI exhibited significant spatial and temporal correlations with vegetation succession, while LSWI behaved in a positive manner with surface water and soil moisture along with the successional stages. In order to take the advantages of both VIs and water index, a composite index of VWR (vegetation water ratio) combining LSWI and EVI or MSAVI was proposed in this paper. This index facilitates the identification of vegetation succession by simply comparing the values of VWR at different stages, and therefore it could track vegetation succession and estimate community spread rate. Additionally, this study presented an attempt of using MODIS datasets to monitor the change of tidal flat elevation, which demonstrated a potential remote sensing application in geodesy of coastal and estuarine areas.     

Combining expert knowledge with machine learning on the basis of fuzzy training

The paper introduces a fuzzy training approach based on nonlinear regularization in an effort to avoid over training. The main idea is to restrict training so that the basic expert knowledge used to build the model is still visible. This is implemented by a new nonlinear regularization approach which can be applied to any kind of training data set. The approach is demonstrated using a large crop yield data set (>4500 field records) for sugar beet collected in agricultural farms over a 14-year period (1976–1989) in East Germany. The software is implemented in SAMT2, free and open source software, using the Python programming language.     

Wheat leaf traits monitoring based on machine learning algorithms and high-resolution satellite imagery

The leaf, which is a crucial indicator for evaluating crop status, plays an important role in plants' functions. Determining and monitoring leaf parameters can facilitate the detection and estimation of crop yield, which is essential for food security. Crop monitoring by remote sensing technology is critical to support crop production, especially over large scales. In this study, we developed a methodology to estimate leaf parameters based entirely on vegetation indices (VIs) from remotely sensed imagery in wheat under different management practices. Therefore, the current study aimed to examine the utility of VIs calculated from the sentinel-2 data in estimating the Leaf area index (LAI) and leaf parameters at wheat farms using machine learning algorithms. Leaf parameters included leaf dry weight (LDW), specific leaf area (SLA) and leaf specific weight (SLW), and machine learning algorithms were SVM (support vector machine), ANN (artificial neural network) and DNN (deep neural network). Leaf parameters were measured at several developmental stages of wheat in two contrasting environments in the southern Iran. The results demonstrated that the DNN algorithm could efficiently predict leaf parameters in the southern Iran with an overall precision of >72%, which assessed the potential of employing DNN to achieve the temporal and spatial distribution data of wheat based on the Sentinel-2 imagery. The validation of the DNN model generally showed high accuracy (R = 0.80, RMSE = 1.19, and MAE = 0.98) between observed and estimated LAI values when this model was used. NDVI was also highly sensitive to wheat LDW and SLA parameters, with a good correlation between field measurements and those predicted by the DNN model from sentinel-2 imagery, with the R values of 0.66 and 0.85, respectively. Further, NDVI and PVI (Perpendicular Vegetation Index) were linearly correlated with SLW across both temporal and spatial scales (R = 0.79). Among VIs considered from sentinel-2 imagery to predict wheat leaf parameters, NDVI was more sensitive than other VIs. This research, thus, indicated that using sentinel-2 data within a DNN model could provide a comparatively precise and robust prediction of leaf parameters and yield valuable insights into crop management with high temporal and spatial accuracy.     

The derivation of sink functions of wheat organs using the GREENLAB model

BACKGROUND AND AIMS: In traditional crop growth models assimilate production and partitioning are described with empirical equations. In the GREENLAB functional-structural model, however, allocation of carbon to different kinds of organs depends on the number and relative sink strengths of growing organs present in the crop architecture. The aim of this study is to generate sink functions of wheat (Triticum aestivum) organs by calibrating the GREENLAB model using a dedicated data set, consisting of time series on the mass of individual organs (the 'target data'). METHODS: An experiment was conducted on spring wheat (Triticum aestivum, 'Minaret'), in a growth chamber from, 2004 to, 2005. Four harvests were made of six plants each to determine the size and mass of individual organs, including the root system, leaf blades, sheaths, internodes and ears of the main stem and different tillers. Leaf status (appearance, expansion, maturity and death) of these 24 plants was recorded. With the structures and mass of organs of four individual sample plants, the GREENLAB model was calibrated using a non-linear least-square-root fitting method, the aim of which was to minimize the difference in mass of the organs between measured data and model output, and to provide the parameter values of the model (the sink strengths of organs of each type, age and tiller order, and two empirical parameters linked to biomass production). KEY RESULTS AND CONCLUSIONS: The masses of all measured organs from one plant from each harvest were fitted simultaneously. With estimated parameters for sink and source functions, the model predicted the mass and size of individual organs at each position of the wheat structure in a mechanistic way. In addition, there was close agreement between experimentally observed and simulated values of leaf area index.     

黄淮海多熟种植农业区作物历遥感检测与时空特征

多熟种植是高强度农业土地利用的重要特征,但由于缺乏在空间和时间上清晰描述农业多熟种植和作物种植历时空分布的数据,使得区域尺度农田生态系统碳动态估计、农田生产力监测与模拟等有很大的不确定性。黄淮海农业区是以冬小麦-夏玉米二熟制为主的我国粮食主产区,冬小麦和夏玉米分别为光合作用途径为C3和C4的作物,已有研究证明如果在估算生态系统生产力时不考虑一年两季作物及其光能利用率的差异则会导致生产力估算结果过低。研究结合农业气象站点地面作物物候观测数据和空间分辨率500m、8d合成的MOD IS时间序列数据,分析研究区二熟制作物的生长过程、物候特征和作物历的空间差异,发展基于EVI和LSWI时间序列曲线检测多熟区各季作物种植历的方法,获取黄淮海农业区空间表述清晰的熟制和各季作物的生长开始与结束时间数据,并应用农业气象站点数据对方法和所获取的作物历数据进行了比较验证。论述的方法和提取的各季作物的作物历时空数据将能够应用于区域尺度农田生产力估算、生物地球化学循环模拟和农业生态系统监测。     

Comparison of Vegetation Indices and Red-edge Parameters for Estimating Grassland Cover from Canopy Reflectance Data

There has been a great deal of Interests in the estimation of grassland biophysical parameters such as percentage of vegetation cover （PVC）, aboveground biomass, and leaf-area index with remote sensing data at the canopy scale. In this paper, the percentage of vegetation cover was estimated from vegetation indices using Moderate Resolution Imaging Spectroradiometer （MODIS） data and red-edge parameters through the first derivative spectrum from in situ hypserspectral reflectance data. Hyperspectral reflectance measurements were made on grasslands in Inner Mongolia, China, using an Analytical Spectral Devices spectroradiometer. Vegetation indices such as the difference, simple ratio, normalized difference, renormalized difference, soil-adjusted and modified soil-adjusted vegetation indices （DVI, RVI, NDVI, RDVI, SAVI L=0.5 end MSAVI2） were calculated from the hyperspectral reflectance of various vegetation covers. The percentage of vegetation cover was estimated using an unsupervised spectral-contextual classifier automatically. Relationships between percentage of vegetation cover and various vegetation indices and red-edge parameters were compared using a linear and second-order polynomial regression. Our analysis indicated that MSAVI2 and RVI yielded more accurate estimations for a wide range of vegetation cover than other vegetation indices and red-edge parameters for the linear and second-order polynomial regression, respectively.     

Temporal and spatial prediction of radiocaesium transfer to food products

A recently developed semi-mechanistic temporal model is used to predict food product radiocaesium activity concentrations using soil characteristics available from spatial soil databases (exchangeable K, pH, percentage clay and percentage organic matter content). A raster database of soil characteristics, radiocaesium deposition, and crop production data has been developed for England and Wales and used to predict the spatial and temporal pattern of food product radiocaesium activity concentrations (Bq/kg). By combining these predictions with spatial data for agricultural production, an area’s output of radiocaesium can also be estimated, we term this flux (Bq/year per unit area). Model predictions have been compared to observed data for radiocaesium contamination of cow milk in regions of England and Wales which received relatively high levels of fallout from the 1986 Chernobyl accident (Gwynedd and Cumbria). The model accounts for 56% and 80% of the observed variation in cow milk activity concentration for Gwynedd and Cumbria, respectively. Illustrative spatial results are presented and suggest that in terms of food product contamination areas in the North and West of England and Wales are those most vulnerable to radiocaesium deposition. When vulnerability is assessed using flux the spatial pattern is more complex and depends upon food product. Received: 26 March 2001 / Accepted: 11 July 2001     

Application of neural computing methods for interpreting phospholipid fatty acid profiles of natural microbial communities

The microbial community compositions of surface and subsurface marine sediments and sediments lining burrows of marine polychaetes and hemichordates from the North Inlet estuary (near Georgetown, S.C. ) were analyzed by comparing ester-linked phospholipid fatty acid (PLFA) profiles with a back-propagating neural network (NN). The NNs were trained to relate PLFA inputs to sediment type outputs (e.g., surface, subsurface, and burrow lining) and worm species (e.g., Notomastus lobatus, Balanoglossus aurantiacus, and Branchyoasychus americana). Sensitivity analysis was used to determine which of the 60 PLFAs significantly contributed to training the NN. The NN architecture was optimized by changing the number of hidden neurons and calculating the cross-validation error between predicted and actual outputs of training and test data. The optimal NN architecture was found to be four hidden neurons with 60-input neurons representing the 60 PLFAs, and four output neurons coding for both sediment types and worm species. Comparison of cross-validation results using NNs and linear discriminant analysis (LDA) revealed that NNs had significantly fewer incorrect classifications (2.7%) than LDA (8.4%). For the NN cross-validation, both sediment type and worm species had 3 incorrect classifications out of 112. For the LDA cross-validation, sediment type and worm species had 7 and 12 incorrect classifications out of 112, respectively. Sensitivity analysis of the trained NNs revealed that 17 fatty acids explained 50% of variability in the data set. These PLFAs were highly different among sediments and burrow types, indicating significant differences in the microbiota.