Welcome to Smart Agriculture 中文
30 June 2021, Volume 3 Issue 2
Topic--Application of Spatial Information Technology in Agriculture
Progress of Agricultural Drought Monitoring and Forecasting Using Satellite Remote Sensing | Open Access
Dong HAN, Pengxin WANG, Yue ZHANG, Huiren TIAN, Xijia ZHOU
2021, 3(2):  1-14.  doi:10.12133/j.smartag.2021.3.2.202104-SA002
Asbtract ( 256 )   HTML ( 51)   PDF (1255KB) ( 166 )
Figures and Tables | References | Related Articles | Metrics

Agricultural drought is a major factor that affects agricultural production. Traditional agricultural drought monitoring is mainly based on meteorological and hydrological data, and although it can provide more accurate drought monitoring results at the point level, there are still limitations in monitoring agricultural drought at the regional scale. The rapid development of remote sensing technology has provided a new mean of monitoring agricultural droughts at the regional scale, especially since the electromagnetic wavelengths sensed by satellite sensors in orbit now cover visible, near-infrared, thermal infrared and microwave wavelengths. It is important to make full use of the rich surface information obtained from satellite remote sensing data for agricultural drought monitoring and forecasting. This paper described the research progress of agricultural drought monitoring based on satellite remote sensing from three aspects: remote sensing index-based method, soil water content method and crop water demand method. The research progress of agricultural drought monitoring based on remote sensing index-based method was elaborated from five aspects: vegetation drought index, temperature drought index, integrated vegetation and temperature drought index, water drought index and microwave drought index; the research progress of agricultural drought monitoring based on soil water content method was elaborated from two aspects: soil water content retrieval based on visible to thermal infrared data and soil water content retrieval based on microwave data; the research progress of agricultural drought monitoring based on crop water demand method was elaborated from two aspects: agricultural drought monitoring based on crop canopy water content retrieval method and crop growth model method. Agricultural drought forecasting is a timeline prediction based on drought monitoring. Based on the summary of the progress of drought monitoring, the research progress of agricultural drought forecasting by the drought index method and the crop growth model method was further briefly described. The existing agricultural drought monitoring methods based on satellite remote sensing were summarized, and its shortcomings were sorted out, and some prospects were put forward. In the future, different remote sensing data sources can be used to combine deep learning methods with crop growth models and based on data assimilation methods to further explore the potential of satellite remote sensing data in the monitoring of agricultural drought dynamics, which can further promote the development of smart agriculture.

Estimating Grain Protein Content of Winter Wheat in Producing Areas Based on Remote Sensing and Meteorological Data | Open Access
Lin WANG, Jian LIANG, Fanyu MENG, Yang MENG, Yongtao ZHANG, Zhenhai LI
2021, 3(2):  15-22.  doi:10.12133/j.smartag.2021.3.2.202103-SA007
Asbtract ( 193 )   HTML ( 27)   PDF (1605KB) ( 114 )
Figures and Tables | References | Related Articles | Metrics

With the rapid development of economy and people's living standards, people's demands for crops have changed from quantity to quality. The rise and rapid development of remote sensing technology provides an effective method for crop monitoring. Accurately predicting wheat quality before harvest is highly desirable to optimize management for farmers, grading harvest and categorized storage for the enterprise, future trading price, and policy planning. In this research, the main producing areas of winter wheat (Henan, Shandong, Hebei, Anhui and Jiangsu provinces) were chosed as the research areas, with collected 898 samples of winter wheat over growing seasons of 2008, 2009 and 2019. A Hierarchical Linear model (HLM) for estimating grain protein content (GPC) of winter wheat at heading-flowering stage was constructed to estimate the GPC of winter wheat in 2019 by using meteorological factors, remote sensing imagery and gluten type of winter wheat, where remote sensing data and gluten type were input variables at the first level of HLM and the meteorological data was used as the second level of HLM. To solve the problem of deviation in interannual and spatial expansion of GPC estimation model, maximum values of Enhanced Vegetation Index (EVI) from April to May calculated by moderate-resolution-imaging spectroradiometer were computed to represent the crop growth status and used in the GPC estimation model. Critical meteorological factors (temperature, precipitation, radiation) and their combinations for GPS estimation were compared and the best estimation model was used in this study. The results showed that the accuracy of GPC considering three meteorological factors performed higher accuracy (Calibrated set: R2 = 0.39, RMSE = 1.04%; Verification set: R2 = 0.43, RMSE = 0.94%) than the others GPC model with two meteorological factors or single meteorological factor. Therefore, three meteorological factors were used as input variables to build a winter wheat GPC forecast model for the regional winter wheat GPC forecast in this research. The GPC estimation model was applied to the GPC remote sensing estimation of the main winter wheat-producing areas, and the GPC prediction map of the main winter wheat producing areas in 2019 was obtained, which could obtain the distribution of winter wheat quality in the Huang-Huai-Hai region. The results of this study could provide data support for subsequent wheat planting regionalization to achieve green, high-yield, high-quality and efficient grain production.

Wheat Lodging Ratio Detection Based on UAS Imagery Coupled with Different Machine Learning and Deep Learning Algorithms | Open Access
Paulo FLORES, Zhao ZHANG
2021, 3(2):  23-34.  doi:10.12133/j.smartag.2021.3.2.202104-SA003
Asbtract ( 269 )   HTML ( 34)   PDF (1857KB) ( 269 )
Figures and Tables | References | Related Articles | Metrics

Wheat lodging is a negative factor affecting yield production. Obtaining timely and accurate wheat lodging information is critical. Using unmanned aerial systems (UASs) images for wheat lodging detection is a relatively new approach, in which researchers usually apply a manual method for dataset generation consisting of plot images. Considering the manual method being inefficient, inaccurate, and subjective, this study developed a new image processing-based approach for automatically generating individual field plot datasets. Images from wheat field trials at three flight heights (15, 46, and 91 m) were collected and analyzed using machine learning (support vector machine, random forest, and K nearest neighbors) and deep learning (ResNet101, GoogLeNet, and VGG16) algorithms to test their performances on detecting levels of wheat lodging percentages: non- (0%), light (<50%), and severe (>50%) lodging. The results indicated that the images collected at 91 m (2.5 cm/pixel) flight height could yield a similar, even slightly higher, detection accuracy over the images collected at 46 m (1.2 cm/pixel) and 15 m (0.4 cm/pixel) UAS mission heights. Comparison of random forest and ResNet101 model results showed that ResNet101 resulted in more satisfactory performance (75% accuracy) with higher accuracy over random forest (71% accuracy). Thus, ResNet101 is a suitable model for wheat lodging ratio detection. This study recommends that UASs images collected at the height of about 91 m (2.5 cm/pixel resolution) coupled with ResNet101 model is a useful and efficient approach for wheat lodging ratio detection.

Identification and Level Discrimination of Waterlogging Stress in Winter Wheat Using Hyperspectral Remote Sensing | Open Access
Feifei YANG, Shengping LIU, Yeping ZHU, Shijuan LI
2021, 3(2):  35-44.  doi:10.12133/j.smartag.2021.3.2.202105-SA001
Asbtract ( 139 )   HTML ( 10)   PDF (1233KB) ( 70 )
Figures and Tables | References | Related Articles | Metrics

The frequent occurrence of waterlogging stress in winter wheat not only seriously affects regional food security and ecological security, but also threatens social and economic stability and sustainable development. In order to identify the waterlogging stress level of winter wheat, a waterlogging stress gradient pot experiment was set up in this research. Three factors were controlled: waterlogging stress level (control, slight waterlogging, severe waterlogging), stress duration (5 days, 10 days, 15 days) and wheat variety (YF4, JM31, JM38). Leaf and canopy hyperspectral data were measured by using ASD Field Spec3 and Gaiasky-mini2 imaging spectrometer, respectively. The data were collected from the first waterlogging day of winter wheat. The sunny and windless weather was selected and measured every 7 days until the wheat was mature. Combined with vegetation index, normalized mean distance and spectral derivative difference entropy, if winter wheat was under waterlogging stress was monitored and stress level was identified. The results showed that: 1) the spectral response characteristics of winter wheat under waterlogging stress changed significantly in RW, RE, NIR and 1650－1800 nm region, which may be due to the sensitivity of these regions to physiological parameters affecting the spectral response characteristics, such as pigment, nutrient, leaf internal structure, etc; 2) the simple ratio pigment index SRPI was the optimal vegetation index for identifying the waterlogging stress of winter wheat. The excellent performance of this vegetation index may come from its extreme sensitivity to the epoxidation state and photosynthetic efficiency of the xanthophyll cycle pigment; 3) the red light absorption valley (RW: 640－680 nm) region was the optimal region for identifying waterlogging stress level. In RW region, waterlogging stress level of winter wheat could be determined by the spectral derivative difference entropy at heading, flowering and filling stages. The greater the level of waterlogging stress, the greater the spectral derivative difference entropy. This may be due to the fact that the RW region was more sensitive to pigment content, and the spectral derivative difference entropy could reduce the effects of spectral noise and background. This study could provide a new method for monitoring waterlogging stress, and would have a good application prospect in the precise prevention and control of waterlogging stress. There are still shortcomings in this study, such as the difference between the pot experiment and the actual field environment, the lack of independent experimental verification, etc. Next research could add pot and field experiments, combine with cross-validation, to further verify the feasibility of this research method.

Comparison of Remote Sensing Estimation Models for Leaf Area Index of Rubber Plantation in Hainan Island | Open Access
Shengpei DAI, Hongxia LUO, Qian ZHENG, Yingying HU, Hailiang LI, Maofen LI, Xuan YU, Bangqian CHEN
2021, 3(2):  45-54.  doi:10.12133/j.smartag.2021.3.2.202106-SA003
Asbtract ( 112 )   HTML ( 5)   PDF (2387KB) ( 49 )
Figures and Tables | References | Related Articles | Metrics

Leaf area index (LAI) is an important index to describe the growth status and canopy structure of vegetation, is of great theoretical and practical significance to quickly obtain LAI of large area vegetation and crops for ecosystem science research and agricultural & forestry production guidance. In this study, the typical tropical crop rubber tree in Hainan Island was selected as the research area, the LAI estimation model of rubber plantation based on satellite remote sensing vegetation indices was constructed, and its spatiotemporal variation was analyzed. The results showed that, compared with correlations between LAI and the indices of normalized difference vegetation index (NDVI), green NDVI (GNDVI), ratio vegetation index (RVI) and wide dynamic range vegetation index (WDRVI), correlations were higher between LAI and the indices of enhanced vegetation index (EVI), soil adjusted vegetation index (SAVI), difference vegetation index (DVI) and modified soil adjusted vegetation index (MSAVI). Among the LAI estimation models based on different vegetation indices (linear, exponential and logarithmic models), the linear estimation model based on EVI index was the best, and its coefficient of determination (R2) was 0.69. The accuracy of LAI estimation model was high. The linear fitting R2 of observed and simulated LAI was 0.67, the root mean square error (RMSE) was 0.16, and the average relative error (RE) was -0.25%. However, there was underestimation in the middle value and overestimation in the high and low value area of LAI. The high LAI values (4.40－6.23) were mainly distributed in Danzhou and Baisha in the west of Hainan Island, the middle LAI values (3.80－4.40) were mainly distributed in Chengmai, Tunchang and Qiongzhong in the middle of Hainan Island, and the low LAI values (2.69－3.80) were mainly distributed in Ding'an, Qionghai, Wanning, Ledong and Sanya in the east and south of Hainan Island. In summary, the linear estimation model for rubber plantation LAI based on EVI index obtained high accuracy, and has good values of popularization and appliance.

Topic--Crop Model and Visualization
Dynamic Simulation of Jujube Tree Growth and Water Use Evaluation Based on the Calibrated WOFOST Model | Open Access
Tiecheng BAI, Tao WANG, Nannan ZHANG
2021, 3(2):  55-67.  doi:10.12133/j.smartag.2021.3.2.202103-SA008
Asbtract ( 289 )   HTML ( 12)   PDF (1931KB) ( 72 )
Figures and Tables | References | Related Articles | Metrics

Irrigation schemes determined based on statistical analysis of field trials are usually only applicable to specific soils and meteorological environments. It is difficult to quantitatively analyze the impact of irrigation strategies on the growth of jujube trees. In order to realize the quantitative analysis of the influence of temperature, light and water resources on the growth of fruit trees, WOrld FOod Studies (WOFOST) model parameters were calibrated to simulate the jujube tree growth and water migration process. Firstly, the observed data obtained from field trials in 2016 and 2017 were used to calibrate the phenology development, initialization, green leaf, CO2 assimilation, dry matter partitioning, respiration, and water use parameters of the WOFOST model. Secondly, the time series of total above-ground biomass, leaf area index (LAI) and soil moisture content in field trials were dynamically simulated, and accuracy verification and analysis were also performed. Finally, the maximum LAI, yield, actual evapotranspiration $(ETa)$ and water use efficiency (WUE) data of 55 orchards were employed to evaluate the performance of the calibrated model at the county scale. The results showed that the coefficient of determination R2 of TAGP simulated in the field test area was between 0.92 and 0.98, and the normalized root mean square error (NRMSE) was between 8.7% and 20.5%, the R2 of simulated LAI ranged from 0.79 to 0.97, and the NRMSE ranged from 8.3% to 21.1%. The R2 of the simulated soil moisture content was between 0.29 and 0.75, and the NRMSE ranged from 4.1% and 6.1%. The model could well simulate the time series of jujube tree growth dynamics and soil moisture content changes. At the county scale, the R2 between the simulated and measured maximum LAI were 0.64 and 0.78, and the NRMSE were 13.3% and 10.7% in 2016 and 2017, respectively. The simulated yield showed R2 value of 0.48 and 0.60, and NRMSE of 12.1% and 11.9%, respectively. RMSE of the simulated versus measured $ETa$ were 36.1 mm (7.9%) and 30.8 mm (7.4%), respectively. The model also showed high WUE simulation accuracy (10%<NRMSE<20%) with RMSE values of 0.23 and 0.28 kg/m3 in 2016 and 2017, respectively. In short, WOFOST model achieved accurate simulation of jujube tree growth and water transport at the field and county scales, which may provide new ideas for the quantitative and mechanism analysis of the coupled effects of soil, weather, irrigation management and jujube tree growth.

Optimum Sowing Date of Winter Wheat in Next 40 Years Based on DSSAT-CERES-Wheat Model | Open Access
Yanan HU, Ju LIANG, Shefang LIANG, Shijuan LI, Yeping ZHU, Yue E
2021, 3(2):  68-76.  doi:10.12133/j.smartag.2021.3.2.202104-SA005
Asbtract ( 122 )   HTML ( 14)   PDF (1378KB) ( 71 )
Figures and Tables | References | Related Articles | Metrics

Climate change requires crop adaptation. Plantint at the suitable date is a key management technology to promote crop yield and address the impact of climate change. Wheat is one of the most important staple crops in China. Huang-Huai-Hai and Jiang-Huai regions are high-quality and high-quantity planting areas for wheat. To deal with the adverse effects of climate change and promote the winter wheat yield in Huang-Huai-Hai and Jiang-Huai regions, the optimum sowing date was identified by creating a wheat simulation with DSSAT CERES-Wheat model. The simulation experiment was designed with 51 management inputs of sowing date and 4 climate scenarios (RCPs) under baseline period (1985－2004) and 40 years in future for three representative stations in the study region. The optimum sowing data of winter wheat was corresponding to the simulation set with highest yield in each site. The characters of changes in climate factors during the growth period and the optimum sowing date among the different period were detected, and the yield increase planted at the optimum sowing date was quantified. The results showed that, in the future, the climate during winter wheat growth period showed a trend of warming and drying would shorten the growth period. The optimum sowing date would be postponed with the rise of temperature, and the decrease of latitude in all periods and under various climate scenarios. Relative to the baseline period, the maximum delay days of the optimal sowing date increased from north to south during 2030s, which were 5 days, 8 days and 13 days at the three representative stations, respectively. The optimum sowing times in 2050s were delayed in different degrees compared with that in 2030s. The largest postponed days at each station was at the RCP8.5 scenario in 2050s. Adopting the management of optimum planting date could mitigate climatic negative effects and was in varying degrees of yield increasing effect at three sites. The smallest increase occurred in Huang-Huai-Hai north region, while Huang-Huai-Hai south region and Jiang-Huai region had the relatively higher yield increasement about 2%－4%. Therefore, the present study demonstrated an effective management of optimum sowing date to promote winter wheat yield under climate change in Huang-Huai-Hai and Jiang-Huai regions.

From Stand to Organ Level—A Trial of Connecting DSSAT and GreenLab Crop Model through Data | Open Access
Xiujuan WANG, Mengzhen KANG, Jing HUA, Philippe DE REFFYE
2021, 3(2):  77-87.  doi:10.12133/j.smartag.2021.3.2.202103-SA006
Asbtract ( 140 )   HTML ( 8)   PDF (2152KB) ( 71 )
Figures and Tables | References | Related Articles | Metrics

Crop models involve complex plant processes, which can be built in different scales of space and time, from molecule, cell, organ, tissue, individual to stand in space and from second to year in time. Based on different research requirements, switching the model scales can make the applicability of the model more extensive and flexible. How to switch the crop model from stand level to organ level is the content of this research. The DSSAT software (stand level) and functional-structural plant model 'GreenLab' (organ level) were chosen to explore the possibility to switch the crop model from stand to organ level. The DSSAT can simulate the growth and development processes of crops in detail according to the growth period by taking the data of weather, soil, crop management, and observational data as input. The GreenLab can simulate the growth and development and their interaction of crops by considering plant structure, and the model parameters can be estimated according to the measurements. In this study, the experimental data contains two parts: the measurements of four maize cultivars with two treatments (irrigated and rainfed) in DSSAT, and the simulations including the weights of leaves, internodes and fruits per day using DSSAT based on the measurements. The simulation results of DSSAT were used to calibrate the parameters of the environmental (E), sink strength (Po), and remobilization (kb and ki) in GreenLab, and to compute the weights of leaves, internodes and fruits for each phytomer. The simulation results of the GreenLab model were compared and analyzed with the experimental data and the simulations of DSSAT. The consistency of calculation results could be an indicator to explore the method of building an interface between different-scale crop models, and to compare the characteristics of different models. The results showed that the GreenLab model could reproduce the simulation data of the DSSAT and the measurement data, including the leaf area index (LAI) and the total weight of the plants, and further could compute the biomass for each organ (leaf, internode and fruit), and the biomass distribution among organs, the biomass production (Q), the demand (D) and the ratio between Q and D during the growth. Therefore, the detailed information of organ growth and development could be reproduced and the 3D structures of plant could be given. Finally, the advantages and application fields of different-scale model integration were discussed.

Information Processing and Decision Making
High-Throughput Dynamic Monitoring Method of Field Maize Seedling | Open Access
Xiaoqing ZHANG, Song SHAO, Xinyu GUO, Jiangchuan FAN
2021, 3(2):  88-99.  doi:10.12133/j.smartag.2021.3.2.202103-SA003
Asbtract ( 142 )   HTML ( 17)   PDF (3369KB) ( 78 )
Figures and Tables | References | Related Articles | Metrics

At present, the dynamic detection and monitoring of maize seedling mainly rely on manual observation, which is time-consuming and laborious, and only small quadrats can be selected to estimate the overall emergence situation. In this research, two kinds of data sources, the high-time-series RGB images obtained by the plant high-throughput phenotypic platform (HTPP) and the RGB images obtained by the unmanned aerial vehicle (UAV) platform, were used to construct the image data set of maize seedling process under different light conditions. Considering the complex background and uneven illumination in the field environment, a residual unit based on the Faster R-CNN was built and ResNet50 was used as a new feature extraction network to optimize Faster R-CNN to realize the detection and counting of maize seedlings in complex field environment. Then, based on the high time series image data obtained by the HTPP, the dynamic continuous monitoring of maize seedlings of different varieties and densities was carried out, and the seedling duration and uniformity of each maize variety were evaluated and analyzed. The experimental results showed that the recognition accuracy of the proposed method was 95.67% in sunny days and 91.36% in cloudy days when it was applied to the phenotypic platform in the field. When applied to the UAV platform to monitor the emergence of maize, the recognition accuracy of sunny and cloudy days was 91.43% and 89.77% respectively. The detection accuracy of the phenotyping platform image was higher, which could meet the needs of automatic detection of maize emergence in actual application scenarios. In order to further verify the robustness and generalization of the model, HTPP was used to obtain time series data, and the dynamic emergence of maize was analyzed. The results showed that the dynamic emergence results obtained by HTPP were consistent with the manual observation results, which shows that the model proposed in this research is robust and generalizable.

Yield Estimation Method of Apple Tree Based on Improved Lightweight YOLOv5 | Open Access
Zhijun LI, Shenghui YANG, Deshuai SHI, Xingxing LIU, Yongjun ZHENG
2021, 3(2):  100-114.  doi:10.12133/j.smartag.2021.3.2.202105-SA005
Asbtract ( 192 )   HTML ( 27)   PDF (3571KB) ( 164 )
Figures and Tables | References | Related Articles | Metrics

Yield estimation of fruit tree is one of the important works in orchard management. In order to improve the accuracy of in-situ yield estimation of apple trees in orchard, a method for the yield estimation of single apple tree, which includes an improved YOLOv5 fruit detection network and a yield fitting network was proposed. The in-situ images of the apples without bags at different periods were acquired by using an unmanned aerial vehicle and Raspberry Pi camera, formed an image sample data set. For dealing with no attention preference and the parameter redundancy in feature extraction, the YOLOv5 network was improved by two approaches: 1) replacing the depth separable convolution, and 2) adding the attention mechanism module, so that the computation cost was decreased. Based on the improvement, the quantity of fruit was estimated and the total area of the bounding box of apples were respectively obtained as output. Then, these results were used as the input of the yield fitting network and actual yields were applied as the output to train the yield fitting network. The final model of fruit tree production estimation was obtained by combining the improved YOLOv5 network and the yield fitting network. Yield estimation experimental results showed that the improved YOLOv5 fruit detection algorithm could improve the recognition accuracy and the degree of lightweight. Compared with the previous algorithm, the detection speed of the algorithm proposed in this research was increased by up to 15.37%, while the mean of average accuracy (mAP) was raised up to 96.79%. The test results based on different data sets showed that the lighting conditions, coloring time and with white cloth in background had a certain impact on the accuracy of the algorithm. In addition, the yield fitting network performed better on predicting the yield of apple trees. The coefficients of determination in the training set and test set were respectively 0.7967 and 0.7982. The prediction accuracy of different yield samples was generally stable. Meanwhile, in terms of the with/without of white cloth in background, the range of relative error of the fruit tree yield measurement model was respectively within 7% and 13%. The yield estimation method of apple tree based on improved lightweight YOLOv5 had good accuracy and effectiveness, which could achieve yield estimation of apples in the natural environment, and would provide a technical reference for intelligent agricultural equipment in modern orchard environment.

EMD-RF-LSTM: Combination Prediction Model of Dissolved Oxygen Concentration in Prawn Culture | Open Access
Hang YIN, Xiangtong LI, Longqin XU, Jingbin LI, Shuangyin LIU, Liang CAO, Dachun FENG, Jianjun GUO, Liqiao LI
2021, 3(2):  115-125.  doi:10.12133/j.smartag.2021.3.2.202106-SA008
Asbtract ( 145 )   HTML ( 18)   PDF (1929KB) ( 69 )
Figures and Tables | References | Related Articles | Metrics

Dissolved oxygen is an important environmental factor for prawn breeding. In order to improve the prediction accuracy of dissolved oxygen concentration in prawn pond, and solve the problem of low prediction accuracy of different frequency domain modal classification after empirical modal decomposition of nonlinear time series data when there are few training samples, an combination prediction model based on empirical mode decomposition (EMD), random forest (RF) and long short term memory neural network (LSTM) was proposed in this research. Firstly, the time series data of prawn breeding dissolved oxygen concentration were decomposed at multiple scales by EMD to obtain a set of stationary intrinsic mode function (IMF). Secondly, with fewer training samples, poor predicts effects on the low-frequency were verified component by LSTM. Then, IMF1－IMF4 were divided into high-frequency components through test results and used for LSTM model. IMF5－IMF7, Rn were divided for RF model, the EMD-RF-LSTM combination model was constructed to improve the prediction accuracy. Modeled low-frequency and high-frequency components IMF using RF and LSTM, then predictions of each component were accumulated and the prediction value of dissolved oxygen of sequence data were got. Finally, the performance of the model was compared with the limit learning machine (ELM), RF, standard LSTM, EMD-ELM and EMD-RF, EMD-LSTM, etc. In the test based on real dataset, the EMD-ELM model contrasted with ELM model, reduced the mean absolute error (MAPE), root mean square error (RMSE) and mean absolute error (MAE) by 30.11%, 29.60% and 32.95%, respectively. The MAPE, RMSE, MAE for the proposed models were 0.0129,0.1156,0.0844, respectively. MAPE decreased by 84.07%, 57.57%, and 49.81% compared with EMD-ELM, EMD-RF and EMD-LSTM, respectively, the prediction accuracy was significantly improved. The results show that the proposed model EMD-RF-LSTM has good prediction performance and generalization ability, which is meets the actual demand of accurate prediction of dissolved oxygen concentration in prawn culture, and can provide reference for the prediction and early warning of prawn pond water quality.

Effect of Growing Season Drought and Flood on Yield of Spring Maize in Three Northeast Provinces of China | Open Access
Weidan WANG, Li SUN, Zhiyuan PEI, Shangjie MA, Yuanyuan CHEN, Juanying SUN, Mo DONG
2021, 3(2):  126-137.  doi:10.12133/j.smartag.2021.3.2.202106-SA004
Asbtract ( 150 )   HTML ( 15)   PDF (1771KB) ( 67 )
Figures and Tables | References | Related Articles | Metrics

Authority in Charge: Ministry of Agriculture and Rural Affairs of the People’s Republic of China

Journal Online
Search by Issue
Search by Key words
Archive By Volume
Smart Agriculture Wechat
Visited