The common practice adopted in previous attempts on Satellite-Derived Bathymetry (SDB) has been to calibrate a single set of coefficients using global regression model. In this study we propose an Adaptive-Geographically Weighted Regression (A-GWR) model that takes into account local factors in determining the regression coefficients. A-GWR model is examined as an effective solution for addressing heterogeneity and could provide better water depth estimates in near-shore region. The study has been carried out for a 30-km stretch and covers 160 km² of a complex near-shore coastal region of Puerto Rico, Northeastern Caribbean Sea. Medium-resolution (Landsat-8) and high-resolution (RapidEye) images were used to estimate water depth. Results demonstrate that the A-GWR model performs well in estimating bathymetry for shallow water depths (1–20 m), showing the correlation coefficient (R) of 0.98 and 0.99, determination coefficient (R²) of 0.95 and 0.99 and Root Mean Square Error (RMSE) of 1.14 and 0.4 m for Landsat-8 and RapidEye, respectively. The data-processing workflow has been entirely implemented in an Open-Source GIS environment and can be easily adopted in other areas.

3D semantic segmentation of point cloud aims at assigning semantic labels to each point by utilizing and respecting the 3D representation of the data. Detailed 3D semantic segmentation of urban areas can assist policymakers, insurance companies, governmental agencies for applications such as urban growth assessment, disaster management, and traffic supervision. The recent proliferation of remote sensing techniques has led to producing high resolution multimodal geospatial data. Nonetheless, currently, only limited technologies are available to fuse the multimodal dataset effectively. Therefore, this paper proposes a novel deep learning-based end-to-end Point-wise LiDAR and Image Multimodal Fusion Network (PMNet) for 3D segmentation of aerial point cloud by fusing aerial image features. PMNet respects basic characteristics of point cloud such as unordered, irregular format and permutation invariance. Notably, multi-view 3D scanned data can also be trained using PMNet since it considers aerial point cloud as a fully 3D representation. The proposed method was applied on two datasets (1) collected from the urban area of Osaka, Japan and (2) from the University of Houston campus, USA and its neighborhood. The quantitative and qualitative evaluation shows that PMNet outperforms other models which use non-fusion and multimodal fusion (observational-level fusion and feature-level fusion) strategies. In addition, the paper demonstrates the improved performance of the proposed model (PMNet) by over-sampling/augmenting the medium and minor classes in order to address the class-imbalance issues.

Chlorophyll content is one of the essential parameters to assess the growth process of the fruit trees. This present study developed a model for estimation of canopy averaged chlorophyll content (CACC) of pear trees using the convolutional auto-encoder (CAE) features of hyperspectral data. This study also demonstrated the inspection of anomaly among the trees by employing multi-dimensional scaling (MDS) on the CAE features and detected outlier trees prior to fit nonlinear regression models. These outlier trees were excluded from the further experiments which helped in improving the prediction performance of CACC. Gaussian process regression (GPR) and support vector regression (SVR) techniques were investigated as nonlinear regression models and used for prediction of CACC. The CAE features were proven to be providing better prediction of CACC when compared with the direct use of hyperspectral bands or vegetation indices as predictors. The CACC prediction performance was improved with the exclusion of the outlier trees during training of the regression models. It was evident from the experiments that GPR could predict the CACC with better accuracy compared to SVR. In addition, the reliability of the tree canopy masks, which were utilized for averaging the features' values for a particular tree, was also evaluated.

Accurate and high resolution bathymetric data is a necessity for a wide range of coastal oceanographic research topics. Active sensing methods, such as ship-based soundings and Light Detection and Ranging (LiDAR), are expensive and time consuming solutions. Therefore, the significance of Satellite-Derived Bathymetry (SDB) has increased in the last ten years due to the availability of multi-constellation, multi-temporal, and multi-resolution remote sensing data as Open Data. Effective SDB algorithms have been proposed by many authors, but there is no ready-to-use software module available in the Geographical Information System (GIS) environment as yet. Hence, this study implements a Geographically Weighted Regression (GWR) based SDB workflow as a Geographic Resources Analysis Support System (GRASS) GIS module (i.image.bathymetry). Several case studies were carried out to examine the performance of the module in multi-constellation and multi-resolution satellite imageries for different study areas. The results indicate a strong correlation between SDB and reference depth. For instance, case study 1 (Puerto Rico, Northeastern Caribbean Sea) has shown an coefficient of determination (R²⁾ of 0.98 and an Root Mean Square Error (RMSE) of 0.61 m, case study 2 (Iwate, Japan) has shown an R² of 0.94 and an RMSE of 1.50 m, and case study 3 (Miyagi, Japan) has shown an R² of 0.93 and an RMSE of 1.65 m. The reference depths were acquired by using LiDAR for case study 1 and an echo-sounder for case studies 2 and 3. Further, the estimated SDB has been used as one of the inputs for the Australian National University and Geoscience Australia (ANUGA) tsunami simulation model. The tsunami simulation results also show close agreement with post-tsunami survey data. The i.mage.bathymetry module developed as a part of this study is made available as an extension for the Open Source GRASS GIS to facilitate wide use and future improvements.

Point cloud is a set of points in 3D space, typically produced by a 3D scanner to capture the 3D representation of a scene. Semantic segmentation of 3D point cloud data where each point is assigned with a semantic class such as building, road, water and so on, has recently gained tremendous attention from data mining researchers and industrial practitioners. Accurate 3D-segmentation results can be used for constructing 3D scene for robotic navigation and assessing the city expansion. Dealing with point cloud data poses a huge challenge of irregular format as points are distributed irregularly unlike 2D pixel of an image or 3D voxel of a 3D model. A number of deep learning architectures have been proposed to model 3D point cloud to perform semantic segmentation. In this paper, we present a new case study of applying three novel deep learning architectures, PointNet, PointCNN and SPGraph, to an outdoor aerial survey point cloud dataset, whose features include intensity and spectral information (RGB). We then compare the results of 3D semantic segmentation from such networks in term of overall accuracy. The result shows that PointNet, PointCNN, and SPGraph achieve 83%, 72.7%, and 83.4% overall accuracy of semantic segmentation, respectively.