Sara Björk

Sara Björk, Stian N. Anfinsen, Michael Kampffmeyer et al.

In prediction of forest parameters with data from remote sensing (RS), regression models have traditionally been trained on a small sample of ground reference data. This paper proposes to impute this sample of true prediction targets with data from an existing RS-based prediction map that we consider as pseudo-targets. This substantially increases the amount of target training data and leverages the use of deep learning (DL) for semi-supervised regression modelling. We use prediction maps constructed from airborne laser scanning (ALS) data to provide accurate pseudo-targets and free data from Sentinel-1's C-band synthetic aperture radar (SAR) as regressors. A modified U-Net architecture is adapted with a selection of different training objectives. We demonstrate that when a judicious combination of loss functions is used, the semi-supervised imputation strategy produces results that surpass traditional ALS-based regression models, even though \sen data are considered as inferior for forest monitoring. These results are consistent for experiments on above-ground biomass prediction in Tanzania and stem volume prediction in Norway, representing a diversity in parameters and forest types that emphasises the robustness of the approach.

Sara Björk, Jonas Nordhaug Myhre, Thomas Haugland Johansen

Full characterization of the spectral behavior of generative models based on neural networks remains an open issue. Recent research has focused heavily on generative adversarial networks and the high-frequency discrepancies between real and generated images. The current solution to avoid this is to either replace transposed convolutions with bilinear up-sampling or add a spectral regularization term in the generator. It is well known that Variational Autoencoders (VAEs) also suffer from these issues. In this work, we propose a simple 2D Fourier transform-based spectral regularization loss for the VAE and show that it can achieve results equal to, or better than, the current state-of-the-art in frequency-aware losses for generative models. In addition, we experiment with altering the up-sampling procedure in the generator network and investigate how it influences the spectral performance of the model. We include experiments on synthetic and real data sets to demonstrate our results.

Sara Björk, Stian Normann Anfinsen, Erik Næsset et al.

This study derives regression models for above-ground biomass (AGB) estimation in miombo woodlands of Tanzania that utilise the high availability and low cost of Sentinel-1 data. The limited forest canopy penetration of C-band SAR sensors along with the sparseness of available ground truth restrict their usefulness in traditional AGB regression models. Therefore, we propose to use AGB predictions based on airborne laser scanning (ALS) data as a surrogate response variable for SAR data. This dramatically increases the available training data and opens for flexible regression models that capture fine-scale AGB dynamics. This becomes a sequential modelling approach, where the first regression stage has linked in situ data to ALS data and produced the AGB prediction map; We perform the subsequent stage, where this map is related to Sentinel-1 data. We develop a traditional, parametric regression model and alternative non-parametric models for this stage. The latter uses a conditional generative adversarial network (cGAN) to translate Sentinel-1 images into ALS-based AGB prediction maps. The convolution filters in the neural networks make them contextual. We compare the sequential models to traditional, non-sequential regression models, all trained on limited AGB ground reference data. Results show that our newly proposed non-sequential Sentinel-1-based regression model performs better quantitatively than the sequential models, but achieves less sensitivity to fine-scale AGB dynamics. The contextual cGAN-based sequential models best reproduce the distribution of ALS-based AGB predictions. They also reach a lower RMSE against in situ AGB data than the parametric sequential model, indicating a potential for further development.

Bjørn Magnus Mathisen, Peter Haro, Bård Hanssen et al.

Decision support systems help decision makers make better decisions in the face of complex decision problems (e.g. investment or policy decisions). Fisheries and Aquaculture is a domain where decision makers face such decisions since they involve factors from many different scientific fields. No systematic overview of literature describing decision support systems and their application in fisheries and aquaculture has been conducted. This paper summarizes scientific literature that describes decision support systems applied to the domain of Fisheries and Aquaculture. We use an established systematic mapping survey method to conduct our literature mapping. Our research questions are: What decision support systems for fisheries and aquaculture exists? What are the most investigated fishery and aquaculture decision support systems topics and how have these changed over time? Do any current DSS for fisheries provide real- time analytics? Do DSSes in Fisheries and Aquaculture build their models using machine learning done on captured and grounded data? The paper then detail how we employ the systematic mapping method in answering these questions. This results in 27 papers being identified as relevant and gives an exposition on the primary methods concluded in the study for designing a decision support system. We provide an analysis of the research done in the studies collected. We discovered that most literature does not consider multiple aspects for multiple stakeholders in their work. In addition we observed that little or no work has been done with real-time analysis in these decision support systems.