Crop Height Estimation of Corn from Multi-Year RADARSAT-2 Polarimetric Observables Using Machine Learning

Abstract

This study presents a demonstration of the applicability of machine learning techniques for the retrieval of crop height in corn fields using space-borne PolSAR (Polarimetric Synthetic Aperture Radar) data. Multi-year RADARSAT-2 C-band data acquired over agricultural areas in Canada, covering the whole corn growing period, are exploited. Two popular machine learning regression methods, i.e., Random Forest Regression (RFR) and Support Vector Regression (SVR) are adopted and evaluated. A set of 27 representative polarimetric parameters are extracted from the PolSAR data and used as input features in the regression models for height estimation. Furthermore, based on the unique capability of the RFR method to determine variable importance contributing to the regression, a smaller number of polarimetric features (6 out of 27 in our study) are selected in the final regression models. Results of our study demonstrate that PolSAR observables can produce corn height estimates with root mean square error (RMSE) around 40–50 cm throughout the growth cycle. The RFR approach shows better overall accuracy in corn height estimation than the SVR method in all tests. The six selected polarimetric features by variable importance ranking can generate better results. This study provides a new perspective on the use of PolSAR data in retrieving agricultural crop height from space.

Crop Monitoring and Classification Using Polarimetric RADARSAT-2 Time-Series Data Across Growing Season: A Case Study in Southwestern Ontario, Canada

“…The crop structural features were also well characterized by their unique scattering patterns using parameters derived from polarimetric decompositions (i.e., FD, CP, and ND). The difference in temporal evolution patterns of scattering mechanisms among crops provided useful information for crop classification, which have been frequently used in crop growth monitoring and classification in previous reports [7,17,27,38,39,41]. Among the polarimetric complex correlation parameters, the correlation coefficient ρ HHVV showed the highest sensitivity to crop growth, which has been proven useful for identification of growth stages [31,36].…”

“…The results using single groups of polarimetric observables showed that polarimetric decompositions (ND, FD, and CP), backscattering coefficients in Pauli and linear polarimetric channels, and correlation coefficients produced the best classification accuracies, with OAs greater than 87%. This is because these polarimetric observables were found to be very sensitive to crop structure and growth parameters [7,17,27,38,39,41]. In previous studies, the common way for constructing the feature set was by stacking all polarimetric observables from various sources [7,17,27,39].…”

“…Moreover, the number and types of selected PolSAR parameters for investigating temporal variation are also limited, such as SAR backscattering and ratios, polarimetric decomposition parameters, polarimetric coherences and phase differences. Meanwhile, fewer polarimetric observables inputting to layer combinations might achieve better results in crop applications [38]. Few efforts have been made in optimal combination of SAR features for crop classification.…”

The Effect of Antecedence on Empirical Model Forecasts of Crop Yield from Observations of Canopy Properties

“…Leaf chlorophyll content directly influences the amount of light captured during photosynthesis and, thus, the amount of energy available for yield production. Remote sensing can also directly measure height [22][23][24], which can be related to crop biomass allometrically and thereby to yield. Taking advantage of new high-resolution satellite data and drone-mounted sensors, it is now possible to produce maps of these yield-related parameters at sub-field scale with known accuracy [25].…”

L-Band SAR Co-Polarized Phase Difference Modeling for Corn Fields

“…Moreover, research on L-band co-polarized phase differences on crops is scarce (e.g., [20]). Most of the research using polarimetric SAR data relied on higher frequencies (C-and X-band) [21][22][23] or multi-polarization intensity-only studies [24]. These shortcomings will be addressed in this manuscript, which turns out to be a novel contribution of this work.…”