ABSTRACT:Remote sensing system mounted on unmanned aerial vehicle (UAV) could provide a complementary means to the conventional satellite and aerial remote sensing solutions especially for the applications of precision agriculture. UAV remote sensing offers a great flexibility to quickly acquire field data in sufficient spatial and spectral resolution at low cost. However a major problem of UAV is the high instability due to the low-end equipments and difficult environment situation, and this leads to image sensor being mostly operated under a highly uncertain configuration. Thus UAV images exhibit considerable derivation in spatial orientation, large geometric and spectral distortion, and low signal-to-noise ratio (SNR). To achieve the objectives of agricultural mapping from UAV, we apply a micro-helicopter UAV with a multiple spectral camera mounted and develop a framework to process UAV images. A very important processing is to generate mosaic image which can be aligned with maps for later GIS integration. With appropriate geometric calibration applied, we first decompose a homography of consecutive image pairs into a rotational component and a simple perspective component, and apply a linear interpolation to the angle of the rotational component, followed by a linear matrix interpolation operator to the perspective component, and this results in an equivalent transformation but ensures a smooth evolution between two images. Lastly to demonstrate the potential of UAV images to precision agriculture application, we perform spectral processing to derive vegetation indices (VIs) maps of crop, and also show the comparison with satellite imagery. Through this paper, we demonstrate that it is highly feasible to generate quantitative mapping products such as crop stress maps from UAV images, and suggest that UAV remote sensing is very valuable for the applications of precision agriculture.

No abstract

A method called a "spectral-spatial-dynamic hierarchical-Bayesian (SSD-HB) model," which can deal with many parameters (such as spectral and weather information all together) by reducing the occurrence of multicollinearity, is proposed. Experiments conducted on soybean yields in Brazil fields with a RapidEye satellite image indicate that the proposed SSD-HB model can predict soybean yield with a higher degree of accuracy than other estimation methods commonly used in remote-sensing applications. In the case of the SSD-HB model, the mean absolute error between estimated yield of the target area and actual yield is 0.28 t/ha, compared to 0.34 t/ha when conventional PLS regression was applied, showing the potential effectiveness of the proposed model.

No abstract

Meta-reinforcement learning (meta-RL) acquires meta-policies that show good performance for tasks in a wide task distribution. However, conventional meta-RL, which learns meta-policies by randomly sampling tasks, has been reported to show meta-overfitting for certain tasks, especially for easy tasks where an agent can easily get high scores.To reduce effects of the meta-overfitting, we considered meta-RL with curriculum-based task sampling. Our method is Robust Meta Reinforcement Learning with Guided Task Sampling (RMRL-GTS), which is an effective method that restricts task sampling based on scores and epochs. We show that in order to achieve robust meta-RL, it is necessary not only to intensively sample tasks with poor scores, but also to restrict and expand the task regions of the tasks to be sampled.