Very little work has been carried out to examine the potential of synthetic aperture radar (SAR) imaging data for growth monitoring, disease monitoring, and yield prediction of oil palm plantations. In this paper, we present a scattering model for oil palm based on the radiative transfer equations, solved iteratively up to the second order, to better understand the backscattering behavior of oil palm canopies for SAR image interpretation. The oil palm canopy is modeled as a multilayer background host medium embedded with discrete scatterers of different shapes, with elliptic disks representing the leaves and cylinders representing the fronds and trunks. We consider the effects of coherent scattering through the dense medium phase and amplitude correction theory, as well as the near-field interactions between closely spaced scatterers through the Fresnel phase and amplitude corrections. Ground-truth data are used as input parameters for the model. The model is used to calculate the backscattering coefficients of oil palm canopies at different stages of growth and for different polarizations, frequencies, and incident angles. Comparisons between model calculations and C-band scatterometer measurements of multipolarization backscattering coefficients of 4-year-old oil palm canopies in Bangi, Malaysia show good agreement. Our model also predicts that SAR images at L-band are more sensitive to changes in the structure of the canopies, particularly that of the fronds, so will be the best frequency for oil palm growth monitoring and disease detection.
14 July 2021 in IEEE Transactions on Geoscience and Remote Sensing
The sea clutter, referred to as the time-varying radar backscatter from the ocean surface, plays a significant role in marine radar development. The ocean’s complex hydrodynamics cause it to exhibit non-Gaussian and nonstationary characteristics, which brings challenges in the sea clutter modeling, especially for establishing its spatial–temporal correlated and coherent model. In this article, a data-driven method based on the Koopman mode decomposition (KMD) is proposed for modeling spatial–temporal correlated complex sea clutter. The method decomposes the coherent sea clutter dynamic behavior in terms of Koopman modes and corresponding temporal patterns. Then, these spatiotemporal patterns are used to construct the sea clutter state over time according to the approximate solution. Furthermore, this proposed state-of-the-art data-driven approach is benchmarked by the measured sea clutter data from intelligent PIXel processing radar (IPIX). It is demonstrated that the proposed approach accurately models the complex sea clutter with actual statistic characteristics, phase information, and spatial–temporal correlations. The mean absolute error (MAE) and root mean square error (RMSE) between the obtained and actual sea clutter are only 0.1817 and 0.2349, respectively. This work offers a practical approach for modeling sea clutter, especially when the spatial–temporal correlation and coherence information is needed.
This paper proposes a novel inverse method based on the deep convolutional neural network (ConvNet) to extract snow’s layer thickness and temperature via passive microwave remote sensing (PMRS). The proposed ConvNet is trained using simulated data obtained through conventional computational electromagnetic methods. Compared with the traditional inverse method, the trained ConvNet can predict the result with higher accuracy. Besides, the proposed method has a strong tolerance for noise. The proposed ConvNet composes three pairs of convolutional and activation layers with one additional fully connected layer to realize regression, i.e., the inversion of snow parameters. The feasibility of the proposed method in learning the inversion of snow parameters is validated by numerical examples. The inversion results indicate that the correlation coefficient () ratio between the proposed ConvNet and conventional methods reaches 4.8, while the ratio for the root mean square error (RMSE) is only 0.18. Hence, the proposed method experiments with a novel path to improve the inversion of passive microwave remote sensing through deep learning approaches.
1 January 2008 in Progress In Electromagnetics Research B
The design and development of synthetic aperture radar\ud (SAR) system for a particular application often requires redesign of software and hardware to optimize the system performance. In addition, evaluations of the performance of existing autofocus and image formation algorithms are required for the SAR system designers to select a most suitable algorithm for a given image quality requirements. This is a time-consuming taskwithout a reconfigurable\ud and comprehensive software package. Thus, a comprehensive SAR integrated simulator and processor software is needed to aid the system designers in optimizing all the system parameters and performance. This paper presents an integrated SAR simulator and processor (iSARSIMP) software package and the performance of three selected SAR autofocus algorithms has been evaluated as examples to demonstrate the usefulness of the iSARSIMP for SAR system designers. In the performance evaluation, simulated and actual SAR raw data were used for further analysis and comparison of the three selected autofocus algorithms
27 June 2022 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
As one of the major cash crops in China, the accurate area and distribution prediction of garlic is of great significance for the adjustment of the extent under cultivation. In this study, to map garlic in the main planting areas of China, an innovative phenology-based automated classification method was developed. The obtained garlic maps were subsequently used to analyze the relationship between areas under cultivation and garlic prices. The details are summarized as follows. First, we established the normalized difference vegetation index time-series remote sensing images to extract winter crop masks. Second, an improved green chlorophyll vegetation index was designed to distinguish garlic from winter crops using Landsat 8 and Sentinel-2 time-series images. Specifically, we extract the garlic planting areas from 2014 to 2021 in our study areas, located in the main planting areas of east-central China. The quantitative evaluations showed that the proposed method delivered an average overall accuracy varying from 89.89% to 98.68% across the entire study areas. Additionally, we analyzed the relationship between the areas in cultivation and the prices of garlic in multiple years. These results indicate that the price of garlic is highly correlated to the planting behavior of farmers, and the size of the planting areas is one of the main factors to the prices of garlic. The results can guide farmers to adjust their planting areas so as to optimize the garlic market and ensure the maximum benefit to farmers.
1 January 2012 in Progress In Electromagnetics Research