Early Classification Method for US Corn and Soybean by Incorporating MODIS-Estimated Phenological Data and Historical Classification Maps in Random-Forest Regression Algorithm

An early crop classification method is functionally required in a near-real-time crop-yield prediction system, especially for upland crops. This study proposes methods to estimate the mixed-pixel ratio of corn, soybean, and other classes within a low-resolution MODIS pixel by coupling MODIS-derived crop phenology information and the past Cropland Data Layer in a random-forest regression algorithm. Verification of the classification accuracy was conducted for the Midwestern United States. The following conclusions are drawn: The use of the random-forest algorithm is effective in estimating the mixed-pixel ratio, which leads to stable classification accuracy; the fusion of historical data and MODIS-derived crop phenology information provides much better crop classification accuracy than when these are used individually; and the input of a longer MODIS data period can improve classification accuracy, especially after day of year 279, because of improved estimation accuracy for the soybean emergence date.

Download Full-text

Estimating and Interpreting Fine-Scale Gridded Population Using Random Forest Regression and Multisource Data

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi9060369 ◽

2020 ◽

Vol 9 (6) ◽

pp. 369

Author(s):

Yun Zhou ◽

Mingguo Ma ◽

Kaifang Shi ◽

Zhenyu Peng

Keyword(s):

Random Forest ◽

Population Distribution ◽

Population Data ◽

Estimation Accuracy ◽

Chinese Government ◽

Random Forest Regression ◽

Regression Techniques ◽

Nighttime Light ◽

Fine Resolution ◽

Impervious Areas

Gridded population results at a fine resolution are important for optimizing the allocation of resources and researching population migration. For example, the data are crucial for epidemic control and natural disaster relief. In this study, the random forest model was applied to multisource data to estimate the population distribution in impervious areas at a 30 m spatial resolution in Chongqing, Southwest China. The community population data from the Chinese government were used to validate the estimation accuracy. Compared with the other regression techniques, the random forest regression method produced more accurate results (R2 = 0.7469, RMSE = 2785.04 and p < 0.01). The points of interest (POIs) data played a more important role in the population estimation than the nighttime light images and natural topographical data, particularly in urban settings. Our results support the wide application of our method in mapping densely populated cities in China and other countries with similar characteristics.

Download Full-text

Estimation of Corn Emergence Date Using UAV Imagery

Transactions of the ASABE ◽

10.13031/trans.14145 ◽

2021 ◽

Vol 64 (4) ◽

pp. 1173-1183

Author(s):

Chin Nee Vong ◽

Stirling A. Stewart ◽

Jianfeng Zhou ◽

Newell R. Kitchen ◽

Kenneth A. Sudduth

Keyword(s):

Machine Learning ◽

Random Forest ◽

Unmanned Aerial Vehicle ◽

Classification Accuracy ◽

Image Features ◽

Shape Features ◽

Size And Shape ◽

Aerial Vehicle ◽

Corn Plants ◽

Emergence Date

HighlightsUAV imagery can be used to characterize newly-emerged corn plants.Size and shape features used in a random forest model are able to predict days after emergence within a 3-day window.Diameter and area were important size features for predicting DAE for the first, second, and third week of emergence.Abstract. Assessing corn (Zea mays L.) emergence uniformity soon after planting is important for relating to grain production and making replanting decisions. Unmanned aerial vehicle (UAV) imagery has been used for determining corn densities at vegetative growth stage 2 (V2) and later, but not as a tool for quantifying emergence date. The objective of this study was to estimate days after corn emergence (DAE) using UAV imagery and a machine learning method. A field experiment was designed with four planting depths to obtain a range of corn emergence dates. UAV imagery was collected during the first, second, and third weeks after emergence. Acquisition height was approximately 5 m above ground level, which resulted in a ground sampling distance of 1.5 mm pixel-1. Seedling size and shape features derived from UAV imagery were used for DAE classification based on a random forest machine learning model. Results showed that 1-day DAE could be distinguished based on image features within the first week after initial corn emergence with a moderate overall classification accuracy of 0.49. However, for the second week and beyond, the overall classification accuracy diminished (0.20 to 0.35). When estimating DAE within a 3-day window (-1 to +1 day), the overall 3-day classification accuracies ranged from 0.54 to 0.88. Diameter, area, and the ratio of major axis length to area were important image features to predict corn DAE. Findings demonstrated that UAV imagery can detect newly-emerged corn plants and estimate their emergence date to assist in assessing emergence uniformity. Additional studies are needed for fine-tuning the image collection procedures and image feature identification to improve accuracy. Keywords: Corn emergence, Image features, Random forest, Unmanned aerial vehicle.

Download Full-text

Estimation of aboveground biomass in North China using Sentinel-1 and 2 datasets

10.5194/egusphere-egu2020-6548 ◽

2020 ◽

Author(s):

Yueting Wang ◽

Xiaoli Zhang

Keyword(s):

Time Series ◽

Random Forest ◽

Aboveground Biomass ◽

North China ◽

Vegetation Indices ◽

Estimation Accuracy ◽

Biophysical Parameters ◽

Random Forest Regression ◽

Multispectral Data ◽

Variables Selection

<p>Forest aboveground biomass (AGB) plays an important role in measuring forest carbon reserves. Accurate mapping AGB is important for monitoring carbon stocks and will contribute to achieve the goal of sustainable development. In this study, we explored the potential of mapping AGB in north China using a three-year monthly time series of Senitinel-1 (S1) and Sentinel-2 (S2) data. The backscattering and indices of SAR S1 combined with spectral reflectance, vegetation indices and biophysical parameters from multispectral S2 imagery were evaluated for AGB prediction in a Random Forest regression.&#160;Three scenarios were conducted with different datasets to determine:&#160;(1) the potential of using S1 and S2 to estimate AGB, (2)&#160;optimal variables selection for AGB mapping, (3)&#160;contribution of time series datasets to improving the accuracy of AGB mapping. Random forest regression was used to develop forest AGB estimation models, which was divided into three types of modeling using only S1, only S2, and a combination of S1 and S2. Compared to S1 (RMSE&#160;= 65.7 Mg/ha), S2 achieved better prediction accuracy (RMSE = 58.4 Mg/ha), although the combination of S1 and S2 time series datasets estimated&#160;the best AGB results (RMSE&#160;= 42.3 Mg/ha).&#160;The research implied that incorporation of SAR and multispetral data considerably improved AGB mapping performance when compared with the use of SAR or multispectral data alone.&#160;This proposed approach provides a new insight in improving the estimation accuracy of forest AGB in north&#160;China.</p>

Download Full-text

POLARIMETRIC SAR DATA FROM SENTINEL-1A APPLIED TO EARLY CROP CLASSIFICATION

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xliii-b3-2020-1039-2020 ◽

2020 ◽

Vol XLIII-B3-2020 ◽

pp. 1039-1046

Author(s):

L. V. Oldoni ◽

V. H. R. Prudente ◽

J. M. F. S. Diniz ◽

N. C. Wiederkehr ◽

I. D. Sanches ◽

...

Keyword(s):

Random Forest ◽

Classification Accuracy ◽

Polarimetric Sar ◽

Early Crop ◽

Development Period ◽

Great Performance ◽

Crop Development ◽

Crop Classification ◽

Sar Data ◽

Multitemporal Data

Abstract. This paper aims to map crops in two Brazilian municipalities, Luís Eduardo Magalhães (LEM) and Campo Verde, using dual-polarimetric Sentinel-1A images. The specific objectives were: (1) to evaluate the accuracy gain in the crop classification using Sentinel-1A multitemporal data backscatter coefficients and ratio (σ0VH, σ0VV and, σ0VH/σ0VV, denominate BS group) in comparison to the addition of polarimetric attributes (σ0VH, σ0VV, σ0VH/σ0VV, H, and α, denominate BP group) and; (2) to assess the accuracy gain in the earliest crop classification, creating new scenarios with the addition of the new SAR data together with the previous images for each date and group (BS and BP) during the crop development. For BS and BP groups, 13 e 10 scenarios were analyzed in LEM and Campo Verde, respectively. For the classification process, we used the Random Forest (RF) algorithm. In the LEM site, the best results for BS and BP groups were equivalent (overall accuracy: ∼82%), while for the Campo Verde site, the classification accuracy for the BP group (overall accuracy: ∼80%) was 2% higher than the BS group. The addition of new images during the crop development period increased the earliest crop classification overall accuracy, stabilizing from mid-February in LEM and mid-December in Campo Verde, after 10 and 8 images, respectively. After these periods, the gain in classification accuracy was small with the addition of new images. In general, our results suggest the backscattering coefficients and polarimetric attributes extracted from the Sentinel-1A imagery exhibited a great performance to discriminate croplands.

Download Full-text

Enhancement of Data Classification Accuracy using Bagging Technique in Random Forest

International Journal of Computer Sciences and Engineering ◽

10.26438/ijcse/v7i8.185188 ◽

2019 ◽

Vol 7 (8) ◽

pp. 185-188

Author(s):

Vikas S. ◽

Thimmaraju S.N.

Keyword(s):

Random Forest ◽

Classification Accuracy ◽

Data Classification

Download Full-text

Perbandingan CART dan Random Forest untuk Deteksi Kanker berbasis Klasifikasi Data Microarray

Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) ◽

10.29207/resti.v4i5.2083 ◽

2020 ◽

Vol 4 (5) ◽

pp. 805-812

Author(s):

Riska Chairunisa ◽

Adiwijaya ◽

Widi Astuti

Keyword(s):

Breast Cancer ◽

Lung Cancer ◽

Ovarian Cancer ◽

Random Forest ◽

Classification And Regression Tree ◽

Classification Method ◽

Discrete Wavelet ◽

Prostate Tumors ◽

Cancer Data ◽

Colon Tumors

Cancer is one of the deadliest diseases in the world with a mortality rate of 57,3% in 2018 in Asia. Therefore, early diagnosis is needed to avoid an increase in mortality caused by cancer. As machine learning develops, cancer gene data can be processed using microarrays for early detection of cancer outbreaks. But the problem that microarray has is the number of attributes that are so numerous that it is necessary to do dimensional reduction. To overcome these problems, this study used dimensions reduction Discrete Wavelet Transform (DWT) with Classification and Regression Tree (CART) and Random Forest (RF) as classification method. The purpose of using these two classification methods is to find out which classification method produces the best performance when combined with the DWT dimension reduction. This research use five microarray data, namely Colon Tumors, Breast Cancer, Lung Cancer, Prostate Tumors and Ovarian Cancer from Kent-Ridge Biomedical Dataset. The best accuracy obtained in this study for breast cancer data were 76,92% with CART-DWT, Colon Tumors 90,1% with RF-DWT, lung cancer 100% with RF-DWT, prostate tumors 95,49% with RF-DWT, and ovarian cancer 100% with RF-DWT. From these results it can be concluded that RF-DWT is better than CART-DWT.

Download Full-text

A Random Forest Regression Model Predicting the Winners of Summer Olympic Events

Proceedings of the 2020 2nd International Conference on Big Data Engineering ◽

10.1145/3404512.3404513 ◽

2020 ◽

Author(s):

Mengjie Jia ◽

Yue Zhao ◽

Furong Chang ◽

Bofeng Zhang ◽

Kenji Yoshigoe

Keyword(s):

Random Forest ◽

Regression Model ◽

Random Forest Regression

Download Full-text

A novel framework for designing a multi-DoF prosthetic wrist control using machine learning

Scientific Reports ◽

10.1038/s41598-021-94449-1 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Chinmay P. Swami ◽

Nicholas Lenhard ◽

Jiyeon Kang

Keyword(s):

Machine Learning ◽

Random Forest ◽

Upper Limb ◽

Daily Living ◽

Machine Learning Algorithms ◽

Data Sets ◽

Random Forest Regression ◽

Prosthetic Devices ◽

Upper Limb Function ◽

The Neural Network

AbstractProsthetic arms can significantly increase the upper limb function of individuals with upper limb loss, however despite the development of various multi-DoF prosthetic arms the rate of prosthesis abandonment is still high. One of the major challenges is to design a multi-DoF controller that has high precision, robustness, and intuitiveness for daily use. The present study demonstrates a novel framework for developing a controller leveraging machine learning algorithms and movement synergies to implement natural control of a 2-DoF prosthetic wrist for activities of daily living (ADL). The data was collected during ADL tasks of ten individuals with a wrist brace emulating the absence of wrist function. Using this data, the neural network classifies the movement and then random forest regression computes the desired velocity of the prosthetic wrist. The models were trained/tested with ADLs where their robustness was tested using cross-validation and holdout data sets. The proposed framework demonstrated high accuracy (F-1 score of 99% for the classifier and Pearson’s correlation of 0.98 for the regression). Additionally, the interpretable nature of random forest regression was used to verify the targeted movement synergies. The present work provides a novel and effective framework to develop an intuitive control for multi-DoF prosthetic devices.

Download Full-text