Improving Crop Mapping by Using Bidirectional Reflectance Distribution Function (BRDF) Signatures with Google Earth Engine

Zhen, Zhijun; Chen, Shengbo; Yin, Tiangang; Gastellu-Etchegorry, Jean-Philippe

doi:10.3390/rs15112761

Cited by 8 publications

(5 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, it has been shown that the use of fitting functions to reconstruct NDVI time series and subsequent GB machine learning make it possible to perform crop mapping based on multi-year data. The high accuracy and productivity of the proposed method determine the prospects for its use in regions with similar cropland structures and crop phenological cycles, including the developing agricultural regions of Northeast China that have large areas of cropland [3,48,49].…”

Section: Discussionmentioning

confidence: 99%

Multi-Year Cropland Mapping Based on Remote Sensing Data: A Case Study for the Khabarovsk Territory, Russia

Dubrovin,

Verkhoturov,

Stepanov

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Cropland mapping using remote sensing data is the basis for effective crop monitoring, crop rotation control, and the detection of irrational land use. Classification using Normalized Difference Vegetation Index (NDVI) time series from multi-year data requires additional time costs, especially when sentinel data are sparse. Approximation by nonlinear functions was proposed to solve this problem. Time series of weekly NDVI composites were plotted using multispectral Sentinel-2 (Level-2A) images at a resolution of 10 m for sites in Khabarovsk District from April to October in the years 2021 and 2022. Missing values due to the lack of suitable images for analysis were recovered using cubic polynomial, Fourier series, and double sinusoidal function approximation. The classes that were considered included crops, namely, soybean, buckwheat, oat, and perennial grasses, and fallow. The mean absolute percentage error (MAPE) of each class fitting was calculated. It was found that Fourier series fitting showed the highest accuracy, with a mean error of 8.2%. Different classifiers, such as the support vector machine (SVM), random forest (RF), and gradient boosting (GB), were comparatively evaluated. The overall accuracy (OA) for the site pixels during the cross-validation (Fourier series restored) was 67.3%, 87.2%, and 85.9% for the SVM, RF, and GB classifiers, respectively. Thus, it was established that the best result in terms of combined accuracy, performance, and limitations in cropland mapping was achieved by composite construction using Fourier series and machine learning using GB. Similar results should be expected in regions with similar cropland structures and crop phenological cycles, including other regions of the Far East.

show abstract

Section: Discussionmentioning

confidence: 99%

Multi-Year Cropland Mapping Based on Remote Sensing Data: A Case Study for the Khabarovsk Territory, Russia

Dubrovin,

Verkhoturov,

Stepanov

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…They explained that the classifiers were ranked according to accuracy, from highest to lowest: RF, CART, SVM, and NB. Finally, they concluded that this study contributes to the development of crop area mapping and the application of multi-angle monitoring satellites [3]. The application of machine learning technology using RF to classify Landsat satellite images in the mountainous terrain of the Himalayas in the western side of the Indian state was presented.…”

Section: Introductionmentioning

confidence: 89%

Comparison of Crop Area Estimation for the Winter Season Using the Google Earth Engine (GEE) Platform and Traditional Method of Supervisor Classification

Amr,

Sheta,

Elwahed

et al. 2024

AJSSPN

View full text Add to dashboard Cite

There is a significant development in the methods and techniques for classifying satellite images during these years. One of the most recently launched methods is using the Google Earth Engine platform for its contribution in providing a huge amount of satellite data with different dates and multiple satellite images sources, with a method of work that is easy to deal with, but it needed to learn the computer language such as Java and Python, which is the method of work depends on writing a script that gives application commands to classify the different satellite image. Therefore, this research aims to apply classification of satellite images using the Google Earth Engine platform for three models Classification and Regression Trees (CART), Support Vector Machine (SVM), and Random Forest (RF), then compare the results calculated from using images classification (Supervised classification: Maximum likelihood) in terms of the degree of accuracy of the classification and matching the results of the classified satellite images in addition. to boundary clarify of the classification methods. This study showed that, using the Google Earth Engine (GEE) platform is considered one of the important methods that facilitated researchers interested in using remote sensing techniques and geographic information systems, due to the availability of this platform of free satellite sources with multiple spatial resolutions, in addition to the presence of a powerful system that collects, processes, and classifies satellite images using different classification models and indexes. But we need to apply these modern techniques at a number of levels, such as at Markas, Governorates, International, and regional levels, to judge their accuracy and suitability for large areas.

show abstract

“…The Bidirectional Reflectance Distribution Function (BRDF) is defined as the ratio of the irradiance in a given outgoing direction to the irradiance in a given incoming direction on that surface element, and this function can be used to characterize the surface reflectance anisotropy [1,2]. Currently, it is mainly applied to radiometric correction of satellite-borne instruments [1], radiometric correction of low-altitude images [3], light utilization, albedo inversion [4,5], crop mapping [6], crop element estimation [7], and estimation of metal surface roughness [8]. The modeling accuracy of the BRDF has directly impacted the application of quantitative remote sensing.…”

Section: Introductionmentioning

confidence: 99%

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Wang,

Li,

Wang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

The Bidirectional Reflectance Distribution Function (BRDF) is a critical spatial distribution parameter in the quantitative research of remote sensing and has a wide range of applications in radiometric correction, elemental inversion, and surface feature estimation. As a new means of BRDF modeling, UAV push-broom hyperspectral imaging is limited by the push-broom imaging method, and the multi-angle information is often difficult to obtain. In addition, the random variation of solar illumination during UAV low-altitude flight makes the irradiance between different push-broom hyperspectral rows and different airstrips inconsistent, which significantly affects the radiometric consistency of BRDF modeling and results in the difficulty of accurately portraying the three-dimensional spatial reflectance distribution in the UAV model. These problems largely impede the application of outdoor BRDF. Based on this, this paper proposes a fast multi-angle information acquisition scheme with a high-accuracy BRDF modeling method considering illumination variations, which mainly involves a lightweight system for BRDF acquisition and three improved BRDF models considering illumination corrections. We adopt multi-rectangular nested flight paths for multi-gray level targets, use multi-mode equipment to acquire spatial illumination changes and multi-angle reflectivity information in real-time, and introduce the illumination correction factor K through data coupling to improve the kernel, Hapke, and RPV models, and, overall, the accuracy of the improved model is increased by 20.83%, 11.11%, and 31.48%, respectively. The results show that our proposed method can acquire multi-angle information quickly and accurately using push-broom hyperspectral imaging, and the improved model eliminates the negative effect of illumination on BRDF modeling. This work is vital for expanding the multi-angle information acquisition pathway and high-efficiency and high-precision outdoor BRDF modeling.

show abstract

Improving Crop Mapping by Using Bidirectional Reflectance Distribution Function (BRDF) Signatures with Google Earth Engine

Cited by 8 publications

References 58 publications

Multi-Year Cropland Mapping Based on Remote Sensing Data: A Case Study for the Khabarovsk Territory, Russia

Multi-Year Cropland Mapping Based on Remote Sensing Data: A Case Study for the Khabarovsk Territory, Russia

Comparison of Crop Area Estimation for the Winter Season Using the Google Earth Engine (GEE) Platform and Traditional Method of Supervisor Classification

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Contact Info

Product

Resources

About