A Survey of Active Learning for Quantifying Vegetation Traits from Terrestrial Earth Observation Data

Abstract

The current exponential increase of spatiotemporally explicit data streams from satellite-based Earth observation missions offers promising opportunities for global vegetation monitoring. Intelligent sampling through active learning (AL) heuristics provides a pathway for fast inference of essential vegetation variables by means of hybrid retrieval approaches, i.e., machine learning regression algorithms trained by radiative transfer model (RTM) simulations. In this study we summarize AL theory and perform a brief systematic literature survey about AL heuristics used in the context of Earth observation regression problems over terrestrial targets. Across all relevant studies it appeared that: (i) retrieval accuracy of AL-optimized training data sets outperformed models trained over large randomly sampled data sets, and (ii) Euclidean distance-based (EBD) diversity method tends to be the most efficient AL technique in terms of accuracy and computational demand. Additionally, a case study is presented based on experimental data employing both uncertainty and diversity AL criteria. Hereby, a a simulated training data base by the PROSAIL-PRO canopy RTM is used to demonstrate the benefit of AL techniques for the estimation of total leaf carotenoid content (Cxc) and leaf water content (Cw). Gaussian process regression (GPR) was incorporated to minimize and optimize the training data set with AL. Training the GPR algorithm on optimally AL-based sampled data sets led to improved variable retrievals compared to training on full data pools, which is further demonstrated on a mapping example. From these findings we can recommend the use of AL-based sub-sampling procedures to select the most informative samples out of large training data pools. This will not only optimize regression accuracy due to exclusion of redundant information, but also speed up processing time and reduce final model size of kernel-based machine learning regression algorithms, such as GPR. With this study we want to encourage further testing and implementation of AL sampling methods for hybrid retrieval workflows. AL can contribute to the solution of regression problems within the framework of operational vegetation monitoring using satellite imaging spectroscopy data, and may strongly facilitate data processing for cloud-computing platforms.

Bridging the Gap Between Remote Sensing and Plant Phenotyping—Challenges and Opportunities for the Next Generation of Sustainable Agriculture

“…Also, deep learning algorithms could be employed, in particular when a large number of different data sets are available and to better describe the highly non-linear relationship between remotely sensed signals and traits of interest. As an additional feature, the quality of training data can be enhanced by implementing active learning heuristics, which recently achieved outstanding results in the estimation of specific traits (Berger et al, 2021 ; Verrelst et al, 2021 ). All in all, these hybrid workflows may become a cornerstone for precision agriculture and an essential element for the development of new breeding strategies (Lammerts van Bueren and Struik, 2017 ).…”

Using Unmanned Aerial Vehicle and Ground-Based RGB Indices to Assess Agronomic Performance of Wheat Landraces and Cultivars in a Mediterranean-Type Environment

“…Some of them have shown a root mean square error (RMSE) ranging from 0.57 to 0.97 t/ha for predicting yield in wheat [18,19]. Other methodologies also use machine-learning regressions, chemometrics, radiative transfer models, photogrammetry, or hybrid approaches to estimate vegetation traits [20][21][22]. On the other hand, far-infrared (thermal) radiation and LIDAR sensors have been respectively used to estimate plant water status [23] and to characterize the architectural features [24].…”

Exploring the Impact of Noise on Hybrid Inversion of PROSAIL RTM on Sentinel-2 Data

“…Furthermore, the presence of noise in satellite data caused by factors such as the atmosphere makes the challenge increasingly complex [25][26][27]. Active learning methods which integrate new samples based on uncertainty or diversity criteria to improve the accuracy of the model in the context of Earth observation regression problems [28], have been shown to enhance retrieval accuracy, which may be due to mitigation of the ill-posedness within hybrid approaches [10], in particular when field data is available [29]. These limitations pose significant challenges even to advanced ML algorithms, and likely contribute to a lack of consensus regarding which algorithm generally performs better.…”