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

2020 ◽  
Vol 12 (21) ◽  
pp. 3504
Author(s):  
Xingrong Li ◽  
Chenghai Yang ◽  
Wenjiang Huang ◽  
Jia Tang ◽  
Yanqin Tian ◽  
...  
Keyword(s):  
Random Forest ◽  
Root Rot ◽  
Vegetation Index ◽  
Texture Features ◽  
Spectral Model ◽  
Spectral Indices ◽  
Texture Model ◽  
Cotton Root ◽  
Cotton Fields ◽  
Sentinel 2

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.

scholarly journals Estimating Above-Ground Biomass of Potato Using Random Forest and Optimized Hyperspectral Indices

2021 ◽  
Vol 13 (12) ◽  
pp. 2339
Author(s):  
Haibo Yang ◽  
Fei Li ◽  
Wei Wang ◽  
Kang Yu
Keyword(s):  
Random Forest ◽  
Vegetation Index ◽  
Tuber Formation ◽  
Growth Stages ◽  
Spectral Indices ◽  
Above Ground Biomass ◽  
Full Spectrum ◽  
Ground Biomass ◽  
Tuber Bulking ◽  
Input Variables

Spectral indices rarely show consistency in estimating crop traits across growth stages; thus, it is critical to simultaneously evaluate a group of spectral variables and select the most informative spectral indices for retrieving crop traits. The objective of this study was to explore the optimal spectral predictors for above-ground biomass (AGB) by applying Random Forest (RF) on three types of spectral predictors: the full spectrum, published spectral indices (Pub-SIs), and optimized spectral indices (Opt-SIs). Canopy hyperspectral reflectance of potato plants, treated with seven nitrogen (N) rates, was obtained during the tuber formation and tuber bulking from 2015 to 2016. Twelve Pub-SIs were selected, and their spectral bands were optimized using band optimization algorithms. Results showed that the Opt-SIs were the best input variables of RF models. Compared to the best empirical model based on Opt-SIs, the Opt-SIs based RF model improved the prediction of AGB, with R2 increased by 6%, 10%, and 16% at the tuber formation, tuber bulking, and for across the two growth stages, respectively. The Opt-SIs can significantly reduce the number of input variables. The optimized Blue nitrogen index (Opt-BNI) and Modified red-edge normalized difference vegetation index (Opt-mND705) combined with an RF model showed the best performance in estimating potato AGB at the tuber formation stage (R2 = 0.88). In the tuber bulking stage, only using optimized Nitrogen planar domain index (Opt-NPDI) as the input variable of the RF model produced satisfactory accuracy in training and testing datasets, with the R2, RMSE, and RE being 0.92, 208.6 kg/ha, and 10.3%, respectively. The Opt-BNI and Double-peak nitrogen index (Opt-NDDA) coupling with an RF model explained 86% of the variations in potato AGB, with the lowest RMSE (262.9 kg/ha) and RE (14.8%) across two growth stages. This study shows that combining the Opt-SIs and RF can greatly enhance the prediction accuracy for crop AGB while significantly reduces collinearity and redundancies of spectral data.

scholarly journals Estimating the Growing Stem Volume of Coniferous Plantations Based on Random Forest Using an Optimized Variable Selection Method

Sensors ◽  
2020 ◽  
Vol 20 (24) ◽  
pp. 7248
Author(s):  
Fugen Jiang ◽  
Mykola Kutia ◽  
Arbi J. Sarkissian ◽  
Hui Lin ◽  
Jiangping Long ◽  
...  
Keyword(s):  
Random Forest ◽  
Variable Selection ◽  
Texture Features ◽  
Estimation Accuracy ◽  
Stem Volume ◽  
Estimation Model ◽  
Red Edge ◽  
Variable Combination ◽  
Coniferous Plantations ◽  
Sentinel 2

Forest growing stem volume (GSV) reflects the richness of forest resources as well as the quality of forest ecosystems. Remote sensing technology enables robust and efficient GSV estimation as it greatly reduces the survey time and cost while facilitating periodic monitoring. Given its red edge bands and a short revisit time period, Sentinel-2 images were selected for the GSV estimation in Wangyedian forest farm, Inner Mongolia, China. The variable combination was shown to significantly affect the accuracy of the estimation model. After extracting spectral variables, texture features, and topographic factors, a stepwise random forest (SRF) method was proposed to select variable combinations and establish random forest regressions (RFR) for GSV estimation. The linear stepwise regression (LSR), Boruta, Variable Selection Using Random Forests (VSURF), and random forest (RF) methods were then used as references for comparison with the proposed SRF for selection of predictors and GSV estimation. Combined with the observed GSV data and the Sentinel-2 images, the distributions of GSV were generated by the RFR models with the variable combinations determined by the LSR, RF, Boruta, VSURF, and SRF. The results show that the texture features of Sentinel-2’s red edge bands can significantly improve the accuracy of GSV estimation. The SRF method can effectively select the optimal variable combination, and the SRF-based model results in the highest estimation accuracy with the decreases of relative root mean square error by 16.4%, 14.4%, 16.3%, and 10.6% compared with those from the LSR-, RF-, Boruta-, and VSURF-based models, respectively. The GSV distribution generated by the SRF-based model matched that of the field observations well. The results of this study are expected to provide a reference for GSV estimation of coniferous plantations.

scholarly journals Monitoring cotton root rot by synthetic Sentinel-2 NDVI time series using improved spatial and temporal data fusion

2018 ◽  
Vol 8 (1) ◽  
Author(s):  
Mingquan Wu ◽  
Chenghai Yang ◽  
Xiaoyu Song ◽  
Wesley Clint Hoffmann ◽  
Wenjiang Huang ◽  
...  
Keyword(s):  
Time Series ◽  
Data Fusion ◽  
Root Rot ◽  
Temporal Data ◽  
Cotton Root ◽  
Ndvi Time Series ◽  
Sentinel 2

scholarly journals Investigating Sentinel 2 Multispectral Imagery Efficiency in Describing Spectral Response of Vineyards Covered with Plastic Sheets

Agronomy ◽  
2020 ◽  
Vol 10 (12) ◽  
pp. 1909
Author(s):  
Enrico Borgogno-Mondino ◽  
Laura de Palma ◽  
Vittorino Novello
Keyword(s):  
Remote Sensing ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Spectral Response ◽  
Table Grape ◽  
Optical Data ◽  
Strong Impact ◽  
Spectral Indices ◽  
Intrinsic Nature ◽  
Sentinel 2

The protection of vineyards with overhead plastic covers is a technique largely applied in table grape growing. As with other crops, remote sensing of vegetation spectral reflectance is a useful tool for improving management even for table grape viticulture. The remote sensing of the spectral signals emitted by vegetation of covered vineyards is currently an open field of investigation, given the intrinsic nature of plastic sheets that can have a strong impact on the reflection from the underlying vegetation. Baring these premises in mind, the aim of the present work was to run preliminary tests on table grape vineyards covered with polyethylene sheets, using Copernicus Sentinel 2 (Level 2A product) free optical data, and compare their spectral response with that of similar uncovered vineyards to assess if a reliable spectral signal is detectable through the plastic cover. Vine phenology, air temperature and shoot growth, were monitored during the 2016 growing cycle. Twenty-four Copernicus Sentinel 2 (S2, Level 2A product) images were used to investigate if, in spite of plastic sheets, vine phenology can be similarly described with and without plastic covers. For this purpose, time series of S2 at-the-ground reflectance calibrated bands and correspondent normalized difference vegetation index (NDVI), modified soil-adjusted vegetation index, version two (MSAVI2) and normalized difference water index (NDWI) spectral indices were obtained and analyzed, comparing the responses of two covered vineyards with different plastic sheets in respect of two uncovered ones. Results demonstrated that no significant limitation (for both bands and spectral indices) was introduced by plastic sheets while monitoring spectral behavior of covered vineyards.

scholarly journals Feature Comparison and Optimization for 30-M Winter Wheat Mapping Based on Landsat-8 and Sentinel-2 Data Using Random Forest Algorithm

2019 ◽  
Vol 11 (5) ◽  
pp. 535 ◽  
Author(s):  
Yuanhuizi He ◽  
Changlin Wang ◽  
Fang Chen ◽  
Huicong Jia ◽  
Dong Liang ◽  
...  
Keyword(s):  
Random Forest ◽  
Winter Wheat ◽  
High Precision ◽  
Agricultural Land ◽  
Global Climate ◽  
Machine Learning Algorithms ◽  
Landsat 8 ◽  
Spectral Indices ◽  
Random Forest Algorithm ◽  
Sentinel 2

Winter wheat cropland is one of the most important agricultural land-cover types affected by the global climate and human activity. Mapping 30-m winter wheat cropland can provide beneficial reference information that is necessary for understanding food security. To date, machine learning algorithms have become an effective tool for the rapid identification of winter wheat at regional scales. Algorithm implementation is based on constructing and selecting many features, which makes feature set optimization an important issue worthy of discussion. In this study, the accurate mapping of winter wheat at 30-m resolution was realized using Landsat-8 Operational Land Imager (OLI), Sentinel-2 Multispectral Imager (MSI) data, and a random forest algorithm. This paper also discusses the optimal combination of features suitable for cropland extraction. The results revealed that: (1) the random forest algorithm provided robust performance using multi-features (MFs), multi-feature subsets (MFSs), and multi-patterns (MPs) as input parameters. Moreover, the highest accuracy (94%) for winter wheat extraction occurred in three zones, including: pure farmland, urban mixed areas, and forest areas. (2) Spectral reflectance and the crop growth period were the most essential features for crop extraction. The MFSs combined with the three to four feature types enabled the high-precision extraction of 30-m winter wheat plots. (3) The extraction accuracy of winter wheat in three zones with multiple geographical environments was affected by certain dominant features, including spectral bands (B), spectral indices (S), and time-phase characteristics (D). Therefore, we can improve the winter wheat mapping accuracy of the three regional types by improving the spectral resolution, constructing effective spectral indices, and enriching vegetation information. The results of this paper can help effectively construct feature sets using the random forest algorithm, thus simplifying the feature construction workload and ensuring high-precision extraction results in future winter wheat mapping research.

scholarly journals Discriminación de trigo y cebada empleando imágenes satelitales ópticas y radar

2021 ◽  
Author(s):  
Mario Fabián Marini
Keyword(s):  
Support Vector Machines ◽  
Maximum Likelihood ◽  
Random Forest ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Buenos Aires ◽  
Support Vector ◽  
Vector Machines ◽  
Sentinel 2

El partido de Coronel Rosales (Buenos Aires, Argentina) se halla localizado dentro de la región pampeana austral, una de las de mayor relevancia agro productiva del país. En este contexto, el conocimiento de la superficie cultivada adquiere significativa importancia para la posterior planificación agrícola y económica. En tal sentido, la discriminación de cultivos mediante teledetección se dificulta cuando se trata de los de ciclo fenológico muy similar, como el trigo y la cebada. En este estudio se realizó una discriminación de dichos cultivos empleando imágenes de Radar de Apertura Sintética (SAR) Sentinel-1A SLC, imágenes ópticas Sentinel-2 y una combinación de ambos tipos de datos. Se incorporaron medidas de coherencia, textura e intensidad de retrodispersión extraídas de los datos SAR durante el ciclo fenológico completo. Sobre cada escena Sentinel-2 se obtuvo el Índice de Diferencia Normalizada de Vegetación (Normalized Difference Vegetation Index - NDVI). Se emplearon tres algoritmos de clasificación: Máxima Verosimilitud (Maximum Likelihood - MLC), Máquinas de Soporte Vectorial (Support Vector Machines - SVM) y Random Forest (RF). Los mejores resultados se obtuvieron al combinar imágenes ópticas y SAR empleando el clasificador RF. La combinación de las retrodispersiones VV y VH junto a la coherencia y la textura de las imágenes SAR, sumada al apilado de NDVI de imágenes ópticas, arrojó los máximos valores de precisión de la clasificación. El valor de F1 fue de 87.27% para el trigo y de 89.20% para la cebada.

scholarly journals Hasil Studi Pola Kebakaran Lahan Gambut melalui Citra Satelit Sentinel-2 dengan Pengimplementasian Machine Learning Metode Random Forest : Kajian Literatur

2021 ◽  
Vol 4 (2) ◽  
pp. 81
Author(s):  
Annisa Rizky Kusuma ◽  
Fauzan Maulana Shodiq ◽  
Muhammad Faris Hazim ◽  
Dany Puguh Laksono
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Ground Water ◽  
Water Level ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Ground Water Level ◽  
Water Index ◽  
Normalized Difference Water Index ◽  
Sentinel 2

Kebakaran lahan gambut merupakan peristiwa yang sulit diprediksi perilakunya. Karakteristik tanah gambut yang kompleks dan faktor-faktor alam lain seperti arah angin, status vegetasi, dan kandungan air membuat kasus ini menjadi salah satu kasus menarik yang masih menjadi objek penelitian yang belum tuntas hingga saat ini. Ketika memasuki musim kemarau kondisi kadar air di dalam tanah gambut akan semakin berkurang, maka potensi terjadinya kebakaran akan semakin tinggi. Pada studi ini dilakukan analisis faktor penyebab kebakaran dengan area cakupan yang luas melalui satelit Sentinel-2. Citra satelit yang diperoleh nantinya akan diolah oleh machine learning untuk memprediksi penyebaran api. Hasil literatur yang telah dilakukan diperoleh bahwa Ground Water Level (GWL), kematangan gambut, suhu, curah hujan dan kelembaban, serta kerapatan vegetasi dapat diidentifikasi melalui perhitungan indeks. Indeks yang digunakan diantaranya indeks Differenced Normalized Difference Vegetation Index (dNDVI) dan Normalized Difference Water Index (NDWI) yang diolah dengan algoritma machine learning metode Random Forest memilki akurasi mencapai 96%.

An attempt to understand corn and winter wheat temporal behavior by means of Sentinel-1 and Sentinel-2 data and its relation with phonological stages.

2021 ◽  
Author(s):  
Marta Pasternak ◽  
Kamila Pawluszek-Filipiak
Keyword(s):  
Remote Sensing ◽  
Winter Wheat ◽  
Strong Correlation ◽  
Vegetation Index ◽  
Radar Data ◽  
Temporal Behavior ◽  
Spectral Indices ◽  
Chlorophyll Index ◽  
Sar Data ◽  
Sentinel 2

<p>Crops are of the fundamental food sources for humanity. Due to the population growth as well as climate change, monitoring of the crops is important to sustain agriculture and conserve natural resources. Development of the remote sensing techniques especially in terms of revisiting time opens new avenues to study crops temporal behaviors from space. Moreover, thanks to the Copernicus program, which guarantees optical as well as radar data to be freely available, there are opportunities to utilize them in an operative way. Additionally, utilization of spectral as well as radar data allows for the synergetic application of both datasets. However, to utilize this data in the operational crop monitoring, it is very important to understand the temporal variations of the remote sensing signal. Therefore, we make an attempt to understand spectral as well as radar remote sensing temporal behavior and its relation with phonological stages.</p><p>For the analysis, 14 cloud-free Sentinel-2 (S-2) acquisitions as well as 34 Sentinel-1 (S-1) acquisitions are utilized. S-2 data were collected with 2A-level while S-1 data was captured in the format of Single Look Complex (SLC) in the Interferometric Wide (IW) swath mode. SLC products consist of complex SAR data preserving phase information which allows studying polarimetric indicators. All remote sensing (spectral as well as SAR) data cover the time period from 04/05/2020 to 07/11/2020. During this time, also 14 field visits were carried out to capture information about phonological stages of corn and wheat according to the BBCH scale (Biologische Bundesanstalt, Bundessortenamt und CHemische Industrie). Additionally, to better understand the temporal behavior of S-1/S-2 signal, weather information from the Institute of Meteorology and Water Management (IMGW) was captured.</p><p>Based on various spectral bands of S-2 data, 12 spectral indices were calculated e.g., GNDVI (Green Normalized Vegetation Index), IRECI (Inverted Red-Edge Chlorophyll Index), MCARI (Modified Chlorophyll Absorption in Reflectance Index), MSAVI (Modified Soil-Adjusted Vegetation Index), MTCI (MERIS Terrestrial Chlorophyll Index), NDVI (Normalized Difference Vegetation Index), PSSRa (Pigment Specific Simple Ratio) and others. After radiometric calibration and the Lee speckle filtering, backscattering coefficients (σ<sub>VV</sub><sup>o</sup> ,σ<sub>VH</sub><sup>o</sup>) of S-1 images were calculated as well as its backscattering ratio (σ<sub>VH</sub><sup>o</sup>/ σ<sub>VV</sub><sup>o</sup>).  All images were then converted from linear to decibel (dB). Additionally, 2 × 2 covariance matrix delivered from S-1 was extracted from the scattering matrix of each SLC image using PolSARpro version 6.0.2 software. After speckle filtration, total scattered power was derived which allows calculating the Shannon Entropy. This value measures the randomness of the scattering within a pixel.</p><p>Time series of many S-2 indices reveal the strong correlation between the development of phenology stages of corn and wheat and the increase of S2 delivered values of spectral indices. However, such a strong correlation cannot be observed within many of S-1 indices. Some of them very poorly indicate the correlation between the development of phenology stages of corn and wheat and increase of S-1 indices values. Additionally, it was observed that values of S1/S2 indices for the same phenology stage very between corn and winter wheat.</p><p> </p>

scholarly journals Identification of Seed Maize Fields With High Spatial Resolution and Multiple Spectral Remote Sensing Using Random Forest Classifier

2020 ◽  
Vol 12 (3) ◽  
pp. 362 ◽  
Author(s):  
Lin Zhang ◽  
Zhe Liu ◽  
Tianwei Ren ◽  
Diyou Liu ◽  
Zhe Ma ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Random Forest ◽  
Vegetation Index ◽  
Seed Quality ◽  
Texture Features ◽  
Kappa Coefficient ◽  
Growth Monitoring ◽  
Landsat 8 ◽  
Wide Range ◽  
Sample Data

Seed maize and common maize plots have different planting patterns and variety types. Identification of seed maize is the basis for seed maize growth monitoring, seed quality and common maize seed supply. In this paper, a random forest (RF) classifier is used to develop an approach for seed maize fields’ identification, using the time series vegetation indexes (VIs) calculated from multispectral data acquired from Landsat 8 and Gaofen 1 satellite (GF-1), field sample data, and texture features of Gaofen 2 satellite (GF-2) panchromatic data. Huocheng and Hutubi County in the Xinjiang Uygur Autonomous Region of China were chosen as study area. The results show that RF performs well with the combination of six VIs (normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), triangle vegetation index (TVI), ratio vegetation index (RVI), normalized difference water index (NDWI) and difference vegetation index (DVI)) and texture features based on a grey-level co-occurrence matrix. The classification based on “spectrum + texture” information has higher overall, user and producer accuracies than that of spectral information alone. Using the “spectrum + texture” method, the overall accuracy of classification in Huocheng County is 95.90%, the Kappa coefficient is 0.92, and the producer accuracy for seed maize fields is 93.91%. The overall accuracy of the classification in Hutubi County is 97.79%, the Kappa coefficient is 0.95, and the producer accuracy for seed maize fields is 97.65%. Therefore, RF classifier inputted with high-resolution remote-sensing image features can distinguish two kinds of planting patterns (seed and common) and varieties types (inbred and hybrid) of maize and can be used to identify and map a wide range of seed maize fields. However, this method requires a large amount of sample data, so how to effectively use and improve it in areas lacking samples needs further research.

scholarly journals Analysis of Combining SAR and Optical Optimal Parameters to Classify Typhoon-Invasion Lodged Rice: A Case Study Using the Random Forest Method

Sensors ◽  
2020 ◽  
Vol 20 (24) ◽  
pp. 7346
Author(s):  
Jinning Wang ◽  
Kun Li ◽  
Yun Shao ◽  
Fengli Zhang ◽  
Zhiyong Wang ◽  
...  
Keyword(s):  
Random Forest ◽  
Large Scale ◽  
Remote Sensing Data ◽  
High Sensitivity ◽  
Spectral Indices ◽  
Optimal Parameters ◽  
Random Forest Classification ◽  
Forest Classification ◽  
Typhoon Event ◽  
Sentinel 2

Lodging, a commonly occurring rice crop disaster, seriously reduces rice quality and production. Monitoring rice lodging after a typhoon event is essential for evaluating yield loss and formulating suitable remedial policies. The availability of Sentinel-1 and Sentinel-2 open-access remote sensing data provides large-scale information with a short revisit time to be freely accessed. Data from these sources have been previously shown to identify lodged crops. In this study, therefore, Sentinel-1 and Sentinel-2 data after a typhoon event were combined to enable monitoring of lodging rice to be quickly undertaken. In this context, the sensitivity of synthetic aperture radar (SAR) features (SF) and spectral indices (SI) extracted from Sentinel-1 and Sentinel-2 to lodged rice were analyzed, and a model was constructed for selecting optimal sensitive parameters for lodging rice (OSPL). OSPL has high sensitivity to lodged rice and strong ability to distinguish lodged rice from healthy rice. After screening, Band 11 (SWIR-1) and Band 12 (SWIR-2) were identified as optimal spectral indices (OSI), and VV, VV + VH and Shannon Entropy were optimal SAR features (OSF). Three classification results of lodging rice were acquired using the Random Forest classification (RFC) method based on OSI, OSF and integrated OSI–OSF stack images, respectively. Results indicate that an overall level of accuracy of 91.29% was achieved with the combination of SAR and optical optimal parameters. The result was 2.91% and 6.05% better than solely using optical or SAR processes, respectively.

Sign in / Sign up

Export Citation Format

Share Document