When sufficient time-series images and training data are unavailable for crop classification, features extracted from convolutional neural network (CNN)-based representative learning may not provide useful information to discriminate crops with similar spectral characteristics, leading to poor classification accuracy. In particular, limited input data are the main obstacles to obtain reliable classification results for early crop mapping. This study investigates the potential of a hybrid classification approach, i.e., CNN-random forest (CNN-RF), in the context of early crop mapping, that combines the automatic feature extraction capability of CNN with the superior discrimination capability of an RF classifier. Two experiments on incremental crop classification with unmanned aerial vehicle images were conducted to compare the performance of CNN-RF with that of CNN and RF with respect to the length of the time-series and training data sizes. When sufficient time-series images and training data were used for the classification, the accuracy of CNN-RF was slightly higher or comparable with that of CNN. In contrast, when fewer images and the smallest training data were used at the early crop growth stage, CNN-RF was substantially beneficial and the overall accuracy increased by maximum 6.7%p and 4.6%p in the two study areas, respectively, compared to CNN. This is attributed to its ability to discriminate crops from features with insufficient information using a more sophisticated classifier. The experimental results demonstrate that CNN-RF is an effective classifier for early crop mapping when only limited input images and training samples are available.

Simulink, an extension of MATLAB, is a graphics-based model development environment for system modeling and simulation. Simulink’s user-friendly features, including block (data process) and arrow (data transfer) objects, a large number of existing blocks, no need to write codes, and a drag and drop interface, provide modelers with an easy development environment. In this study, a Tank model was developed using Simulink and applied to a rainfall-runoff simulation for a study watershed to demonstrate the potential of Simulink as a tool for hydrological analysis. In the example given here, the Tank model was extended by two sub-modules representing evapotranspiration and storage-runoff distribution. In addition, model pre- and post-processing, such as input data preparation and results plotting, was carried out in MATLAB. Moreover, model parameters were calibrated using MATLAB optimization tools without any additional programming for linking the calibration algorithms and the model. The graphical representation utilized in the Simulink version of the Tank model helped us to understand the hydrological interactions described in the model, and the modular structure of the program facilitated the addition of new modules and the modification of existing modules as needed. From the study, we found that Simulink could be a useful and convenient environment for hydrological analysis and model development.

To improve the accuracy of classification with a small amount of training data, this paper presents a self-learning approach that defines class labels from sequential patterns using a series of past land-cover maps. By stacking past land-cover maps, unique sequence rule information from sequential change patterns of land-covers is first generated, and a rule-based class label image is then prepared for a given time. After the most informative pixels with high uncertainty are selected from the initial classification, rule-based class labels are assigned to the selected pixels. These newly labeled pixels are added to training data, which then undergo an iterative classification process until a stopping criterion is reached. Time-series MODIS NDVI data sets and cropland data layers (CDLs) from the past five years are used for the classification of various crop types in Kansas. From the experiment results, it is found that once the rule-based labels are derived from past CDLs, the labeled informative pixels could be properly defined without analyst intervention. Regardless of different combinations of past CDLs, adding these labeled informative pixels to training data increased classification accuracy and the maximum improvement of 8.34 percentage points in overall accuracy was achieved when using three CDLs, compared to the initial classification result using a small amount of training data. Using more than three consecutive CDLs showed slightly better classification accuracy than when using two CDLs (minimum and maximum increases were 1.56 and 2.82 percentage points, respectively). From a practical viewpoint, using three or four CDLs was the best choice for this study area. Based on these experiment results, the presented approach could be applied effectively to areas with insufficient training data but access to past land-cover maps. However, further consideration should be given to select the optimal number of past land-cover maps and reduce the impact of errors of rule-based labels.

It is important to be able to predict the yield and monitor the growth conditions of crops in the field to increase productivity. One way to assess field-based geospatial crop productivity is by integrating a crop model with a remote-controlled aerial system (RAS). The objective of this study was to simulate spatiotemporal barley growth and yield based on the development of a crop-modeling system integrated with RAS-based remote sensing images. We performed field experiments to obtain ground truth data and RAS images of crop growth conditions and yields at Chonnam National University (CNU), Gwangju, South Korea in 2018, and at Gyeongsang National University (GNU), Jinju, South Gyeongsang, South Korea in 2018 and 2019. In model calibration, there was no significant difference (p = 0.12) between the simulated barley yields and measured yields, based on a two-sample t-test at CNU in 2018. In model validation, there was no significant difference between simulated yields and measured yields at p = 0.98 and 0.76, according to two-sample t-tests at GNU in 2018 and 2019, respectively. The remote sensing-integrated crop model accurately reproduced geospatial variations in barley yield and growth variables. The results demonstrate that the crop modeling approach is useful for monitoring at-field barley conditions.

Availability of high resolution remote sensing images along with advances in deep learning has enabled timely generation of crop type maps. However, scaling such maps to large areas is often hindered by data occlusion due to clouds. In this paper, we propose an ensemble of image segmentation models, based on attention U-Net architecture and optimized using Focal Trvesky loss, to scale crop-type mapping by tackling the missing data problem. We trained models on Landsat 8 and Sentinel 2 surface reflectance data and created a bagged ensemble of those trained models to generate wall-to-wall crop type maps of the entire US corn-belt. We also demonstrate the effectiveness of our approach in generating pre-harvest wall-to-wall crop type maps. Our proposed method obtains a competitive 92% overall accuracy for crop type mapping without requiring the availability of the entire in-season time-series of remote sensing images over the entire mapping region.