Evaluation of Sentinel-2A Satellite Imagery for Mapping Cotton Root Rot

A Comprehensive Review of High Throughput Phenotyping and Machine Learning for Plant Stress Phenotyping

During the last decade, there has been rapid adoption of ground and aerial platforms with multiple sensors for phenotyping various biotic and abiotic stresses throughout the developmental stages of the crop plant. High throughput phenotyping (HTP) involves the application of these tools to phenotype the plants and can vary from ground-based imaging to aerial phenotyping to remote sensing. Adoption of these HTP tools has tried to reduce the phenotyping bottleneck in breeding programs and help to increase the pace of genetic gain. More specifically, several root phenotyping tools are discussed to study the plant's hidden half and an area long neglected. However, the use of these HTP technologies produces big data sets that impede the inference from those datasets. Machine learning and deep learning provide an alternative opportunity for the extraction of useful information for making conclusions. These are interdisciplinary approaches for data analysis using probability, statistics, classification, regression, decision theory, data visualization, and neural networks to relate information extracted with the phenotypes obtained. These techniques use feature extraction, identification, classification, and prediction criteria to identify pertinent data for use in plant breeding and pathology activities. This review focuses on the recent findings where machine learning and deep learning approaches have been used for plant stress phenotyping with data being collected using various HTP platforms. We have provided a comprehensive overview of different machine learning and deep learning tools available with their potential advantages and pitfalls. Overall, this review provides an avenue for studying various HTP platforms with particular emphasis on using the machine learning and deep learning tools for drawing legitimate conclusions. Finally, we propose the conceptual challenges being faced and provide insights on future perspectives for managing those issues.

Wood Decay Detection in Norway Spruce Forests Based on Airborne Hyperspectral and ALS Data

Wood decay caused by pathogenic fungi in Norway spruce forests causes severe economic losses in the forestry sector, and currently no efficient methods exist to detect infected trees. The detection of wood decay could potentially lead to improvements in forest management and could help in reducing economic losses. In this study, airborne hyperspectral data were used to detect the presence of wood decay in the trees in two forest areas located in Etnedal (dataset I) and Gran (dataset II) municipalities, in southern Norway. The hyperspectral data used consisted of images acquired by two sensors operating in the VNIR and SWIR parts of the spectrum. Corresponding ground reference data were collected in Etnedal using a cut-to-length harvester while in Gran, field measurements were collected manually. Airborne laser scanning (ALS) data were used to detect the individual tree crowns (ITCs) in both sites. Different approaches to deal with pixels inside each ITC were considered: in particular, pixels were either aggregated to a unique value per ITC (i.e., mean, weighted mean, median, centermost pixel) or analyzed in an unaggregated way. Multiple classification methods were explored to predict rot presence: logistic regression, feed forward neural networks, and convolutional neural networks. The results showed that wood decay could be detected, even if with accuracy varying among the two datasets. The best results on the Etnedal dataset were obtained using a convolution neural network with the first five components of a principal component analysis as input (OA = 65.5%), while on the Gran dataset, the best result was obtained using LASSO with logistic regression and data aggregated using the weighted mean (OA = 61.4%). In general, the differences among aggregated and unaggregated data were small.

Identification of Cotton Leaf Lesions Using Deep Learning Techniques

The use of deep learning models to identify lesions on cotton leaves on the basis of images of the crop in the field is proposed in this article. Cultivated in most of the world, cotton is one of the economically most important agricultural crops. Its cultivation in tropical regions has made it the target of a wide spectrum of agricultural pests and diseases, and efficient solutions are required. Moreover, the symptoms of the main pests and diseases cannot be differentiated in the initial stages, and the correct identification of a lesion can be difficult for the producer. To help resolve the problem, the present research provides a solution based on deep learning in the screening of cotton leaves which makes it possible to monitor the health of the cotton crop and make better decisions for its management. With the learning models GoogleNet and Resnet50 using convolutional neural networks, a precision of 86.6% and 89.2%, respectively, was obtained. Compared with traditional approaches for the processing of images such as support vector machines (SVM), Closest k-neighbors (KNN), artificial neural networks (ANN) and neuro-fuzzy (NFC), the convolutional neural networks proved to be up to 25% more precise, suggesting that this method can contribute to a more rapid and reliable inspection of the plants growing in the field.

Identification of Cotton Root Rot by Multifeature Selection from Sentinel-2 Images Using Random Forest

Cotton root rot is a destructive cotton disease and significantly affects cotton quality and yield, and accurate identification of its distribution within fields is critical for cotton growers to control the disease effectively. In this study, Sentinel-2 images were used to explore the feasibility of creating classification maps and prescription maps for site-specific fungicide application. Eight cotton fields with different levels of root rot were selected and random forest (RF) was used to identify the optimal spectral indices and texture features of the Sentinel-2 images. Five optimal spectral indices (plant senescence reflectance index (PSRI), normalized difference vegetation index (NDVI), normalized difference water index (NDWI1), moisture stressed index (MSI), and renormalized difference vegetation index (RDVI)) and seven optimal texture features (Contrast 1, Dissimilarity 1, Entory 2, Mean 1, Variance 1, Homogeneity 1, and Second moment 2) were identified. Three binary logistic regression (BLR) models, including a spectral model, a texture model, and a spectral-texture model, were constructed for cotton root rot classification and prescription map creation. The results were compared with classification maps and prescription maps based on airborne imagery. Accuracy assessment showed that the accuracies of the classification maps for the spectral, texture, and spectral-texture models were 92.95%, 84.81%, and 91.87%, respectively, and the accuracies of the prescription maps for the three respective models were 90.83%, 87.14%, and 91.40%. These results confirmed that it was feasible to identify cotton root rot and create prescription maps using different features of Sentinel-2 imagery. The addition of texture features had little effect on the overall accuracy, but it could improve the ability to identify root rot areas. The producer’s accuracy (PA) for infested cotton in the classification maps for the texture model and the spectral-texture model was 2.82% and 1.07% higher, respectively, than that of the spectral model, and the PA for treatment zones in the prescription maps for the two respective models was 8.6% and 8.22% higher than that of the spectral model. Results based on the eight cotton fields showed that the spectral model was appropriate for the cotton fields with relatively severe infestation and the spectral-texture model was more appropriate for the cotton fields with low or moderate infestation.

A Plant-by-Plant Method to Identify and Treat Cotton Root Rot Based on UAV Remote Sensing

Cotton root rot (CRR), caused by the fungus Phymatotrichopsis omnivora, is a destructive cotton disease that mainly affects the crop in Texas. Flutriafol fungicide applied at or soon after planting has been proven effective at protecting cotton plants from being infected by CRR. Previous research has indicated that CRR will reoccur in the same regions of a field as in past years. CRR-infected plants can be detected with aerial remote sensing (RS). As unmanned aerial vehicles (UAVs) have been introduced into agricultural RS, the spatial resolution of farm images has increased significantly, making plant-by-plant (PBP) CRR classification possible. An unsupervised classification algorithm, PBP, based on the Superpixel concept, was developed to delineate CRR-infested areas at roughly the single-plant level. Five-band multispectral data were collected with a UAV to test these methods. The results indicated that the single-plant level classification achieved overall accuracy as high as 95.94%. Compared to regional classifications, PBP classification performed better in overall accuracy, kappa coefficient, errors of commission, and errors of omission. The single-plant fungicide application was also effective in preventing CRR.

Estimating Wheat Grain Yield Using Sentinel-2 Imagery and Exploring Topographic Features and Rainfall Effects on Wheat Performance in Navarre, Spain

Reliable methods for estimating wheat grain yield before harvest could help improve farm management and, if applied on a regional level, also help identify spatial factors that influence yield. Regional grain yield can be estimated using conventional methods, but the typical process is complex and labor-intensive. Here we describe the development of a streamlined approach using publicly accessible agricultural data, field-level yield, and remote sensing data from Sentinel-2 satellite to estimate regional wheat grain yield. We validated our method on wheat croplands in Navarre in northern Spain, which features heterogeneous topography and rainfall. First, this study developed stepwise multilinear equations to estimate grain yield based on various vegetation indices, which were measured at various phenological stages in order to determine the optimal timings. Second, the most suitable model was used to estimate grain yield in wheat parcels mapped from Sentinel-2 satellite images. We used a supervised pixel-based random forest classification and the estimates were compared to government-published post-harvest yield statistics. When tested, the model achieved an R2 of 0.83 in predicting grain yield at field level. The wheat parcels were mapped with an accuracy close to 86% for both overall accuracy and compared to official statistics. Third, the validated model was used to explore potential relationships of the mapped per-parcel grain yield estimation with topographic features and rainfall by using geographically weighted regressions. Topographic features and rainfall together accounted for an average for 11 to 20% of the observed spatial variation in grain yield in Navarre. These results highlight the ability of our method for estimating wheat grain yield before harvest and determining spatial factors that influence yield at the regional scale.

Automatic Estimation of Crop Disease Severity Levels Based on Vegetation Index Normalization

The timely monitoring of crop disease development is very important for precision agriculture applications. Remote sensing-based vegetation indices (VIs) can be good indicators of crop disease severity, but current methods are mainly dependent on manual ground survey results. Based on VI normalization, an automated crop disease severity grading method without the use of ground surveys was proposed in this study. This technique was applied to two cotton fields infested with different levels of cotton root rot in south Texas in the United States, where airborne hyperspectral imagery was collected. Six typical VIs were calculated from the hyperspectral imagery and their histograms indicated that VI normalization could eliminate the influences of variable field conditions and the VI value range variations, allowing a potentially broader scope of application. According to the analysis of the obtained results from the spectral dimension, spatial dimension and descriptive statistics, the disease grading results were in general agreement with previous ground survey results, proving the validity of the disease severity grading method. Although satisfactory results could be achieved from different types of VI, there is still room for further improvement through the exploration of more VIs. With the advantages of independence of ground surveys and potential universal applicability, the newly proposed crop disease grading method will be of great significance for crop disease monitoring over large geographical areas.

Remote Sensing for Precision Agriculture: Sentinel-2 Improved Features and Applications

The use of satellites to monitor crops and support their management is gathering increasing attention. The improved temporal, spatial, and spectral resolution of the European Space Agency (ESA) launched Sentinel-2 A + B twin platform is paving the way to their popularization in precision agriculture. Besides the Sentinel-2 A + B constellation technical features the open-access nature of the information they generate, and the available support software are a significant improvement for agricultural monitoring. This paper was motivated by the challenges faced by researchers and agrarian institutions entering this field; it aims to frame remote sensing principles and Sentinel-2 applications in agriculture. Thus, we reviewed the features and uses of Sentinel-2 in precision agriculture, including abiotic and biotic stress detection, and agricultural management. We also compared the panoply of satellites currently in use for land remote sensing that are relevant for agriculture to the Sentinel-2 A + B constellation features. Contrasted with previous satellite image systems, the Sentinel-2 A + B twin platform has dramatically increased the capabilities for agricultural monitoring and crop management worldwide. Regarding crop stress monitoring, Sentinel-2 capacities for abiotic and biotic stresses detection represent a great step forward in many ways though not without its limitations; therefore, combinations of field data and different remote sensing techniques may still be needed. We conclude that Sentinel-2 has a wide range of useful applications in agriculture, yet still with room for further improvements. Current and future ways that Sentinel-2 can be utilized are also discussed.

Land Use/Land Cover Mapping Using Multitemporal Sentinel-2 Imagery and Four Classification Methods—A Case Study from Dak Nong, Vietnam

Information on land use and land cover (LULC) including forest cover is important for the development of strategies for land planning and management. Satellite remotely sensed data of varying resolutions have been an unmatched source of such information that can be used to produce estimates with a greater degree of confidence than traditional inventory estimates. However, use of these data has always been a challenge in tropical regions owing to the complexity of the biophysical environment, clouds, and haze, and atmospheric moisture content, all of which impede accurate LULC classification. We tested a parametric classifier (logistic regression) and three non-parametric machine learning classifiers (improved k-nearest neighbors, random forests, and support vector machine) for classification of multi-temporal Sentinel 2 satellite imagery into LULC categories in Dak Nong province, Vietnam. A total of 446 images, 235 from the year 2017 and 211 from the year 2018, were pre-processed to gain high quality images for mapping LULC in the 6516 km2 study area. The Sentinel 2 images were tested and classified separately for four temporal periods: (i) dry season, (ii) rainy season, (iii) the entirety of the year 2017, and (iv) the combination of dry and rainy seasons. Eleven different LULC classes were discriminated of which five were forest classes. For each combination of temporal image set and classifier, a confusion matrix was constructed using independent reference data and pixel classifications, and the area on the ground of each class was estimated. For overall temporal periods and classifiers, overall accuracy ranged from 63.9% to 80.3%, and the Kappa coefficient ranged from 0.611 to 0.813. Area estimates for individual classes ranged from 70 km2 (1% of the study area) to 2200 km2 (34% of the study area) with greater uncertainties for smaller classes.

Automatic Classification of Cotton Root Rot Disease Based on UAV Remote Sensing

Cotton root rot (CRR) is a persistent soilborne fungal disease that is devastating to cotton in the southwestern U.S. and Mexico. Research has shown that CRR can be prevented or at least mitigated by applying a fungicide at planting, but the fungicide should be applied precisely to minimize the quantity of product used and the treatment cost. The CRR-infested areas within a field are consistent from year to year, so it is possible to apply the fungicide only at locations where CRR is manifest, thus minimizing the amount of fungicide applied across the field. Previous studies have shown that remote sensing (RS) from manned aircraft is an effective means of delineating CRR-infested field areas. Applying various classification methods to moderate-resolution (1.0 m/pixel) RS images has recently become the conventional way to delineate CRR-infested areas. In this research, an unmanned aerial vehicle (UAV) was used to collect high-resolution remote sensing (RS) images in three Texas fields known to be infested with CRR. Supervised, unsupervised, and combined unsupervised classification methods were evaluated for differentiating CRR from healthy zones of cotton plants. Two new automated classification methods that take advantage of the high resolution inherent in UAV RS images were also evaluated. The results indicated that the new automated methods were up to 8.89% better than conventional classification methods in overall accuracy. One of these new methods, an automated method combining k-means segmentation and morphological opening and closing, provided the best results, with overall accuracy of 88.5% and the lowest errors of omission (11.44%) and commission (16.13%) of all methods considered.