The retrieval of chlorophyll-a (Chl-a) concentrations relies on empirical or analytical analyses, which generally experience difficulties from the diversity of inland waters in statistical analyses and the complexity of radiative transfer equations in analytical analyses, respectively. Previous studies proposed the utilization of artificial neural networks (ANNs) to alleviate these problems. However, ANNs do not consider the problem of insufficient in situ samples during model training, and they do not fully utilize the spatial and spectral information of remote sensing images in neural networks. In this study, a two-stage training is introduced to address the problem regarding sample insufficiency. The neural network is pretrained using the samples derived from an existing Chl-a concentration model in the first stage, and the pretrained model is refined with in situ samples in the second stage. A novel convolutional neural network for Chl-a concentration retrieval called WaterNet is proposed which utilizes both spectral and spatial information of remote sensing images. In addition, an end-to-end structure that integrates feature extraction, band expansion, and Chl-a estimation into the neural network leads to an efficient and effective Chl-a concentration retrieval. In experiments, Sentinel-3 images with the same acquisition days of in situ measurements over Laguna Lake in the Philippines were used to train and evaluate WaterNet. The quantitative analyses show that the two-stage training is more likely than the one-stage training to reach the global optimum in the optimization, and WaterNet with two-stage training outperforms, in terms of estimation accuracy, related ANN-based and band-combination-based Chl-a concentration models.

An increasing number of three-dimensional (3D) building models are being made available on Web-based model-sharing platforms. Motivated by the concept of data reuse, an encoding approach is proposed for 3D building model retrieval using point clouds acquired by airborne light detection and ranging (LiDAR) systems. To encode LiDAR point clouds with sparse, noisy, and incomplete sampling, we introduce a novel encoding scheme based on a set of low-frequency spherical harmonic basis functions. These functions provide compact representation and ease the encoding difficulty coming from inherent noises of point clouds. Additionally, a data filling and resampling technique is proposed to solve the aliasing problem caused by the sparse and incomplete sampling of point clouds. Qualitative and quantitative analyses of LiDAR data show a clear superiority of the proposed method over related methods. A cyber campus generated by retrieving 3D building models with airborne LiDAR point clouds demonstrates the feasibility of the proposed method.

The spatial heterogeneity and nonlinearity exhibited by bio-optical relationships in turbid inland waters complicate the retrieval of chlorophyll-a (Chl-a) concentration from multispectral satellite images. Most studies achieved satisfactory Chl-a estimation and focused solely on the spectral regions from near-infrared (NIR) to red spectral bands. However, the optical complexity of turbid waters may vary with locations and seasons, which renders the selection of spectral bands challenging. Accordingly, this study proposes an optimization process utilizing available spectral models to achieve optimal Chl-a retrieval. The method begins with the generation of a set of feature candidates, followed by candidate selection and optimization. Each candidate links to a Chl-a estimation model, including two-band, three-band, and normalized different chlorophyll index models. Moreover, a set of selected candidates using available spectral bands implies an optimal composition of estimation models, which results in an optimal Chl-a estimation. Remote sensing images and in situ Chl-a measurements in Lake Kasumigaura, Japan, are analyzed quantitatively and qualitatively to evaluate the proposed method. Results indicate that the model outperforms related Chl-a estimation models. The root-mean-squared errors of the Chl-a concentration obtained by the resulting model (OptiM-3) improve from 11.95 $\mathrm{mg}·{\mathrm{m}}^{-3}$ to 6.37 $\mathrm{mg}·{\mathrm{m}}^{-3}$ , and the Pearson’s correlation coefficients between the predicted and in situ Chl- $\mathrm{a}$ improve from 0.56 to 0.89.

Radiometric normalization is a fundamental preprocessing step for the analysis of multi-temporal and multi-sensor satellite images. Some of previous radiometric normalization studies addressed the elimination of spectral differences between bitemporal or cross-sensor images by performing canonical correlation analysis (CCA) or its generalization called kernel CCA (KCCA). The KCCA-based normalization methods, which assume that the relation between at-sensor radiances is spatially nonlinear, can successfully extract pseudo-invariant features (PIFs) for bitemporal images with nonlinear differences caused by seasonal variation and for bitemporal images acquired by different sensors with different spectral centers and bandwidths of spectral bands. However, these methods did not consider the problems of high computation and storage costs induced by the huge size of kernel matrices and overfitting problem caused by non-linear regression, making the methods infeasible for satellite images of large sizes. To solve these problems, a hybrid CCA with hybrid regression model is proposed, which can preserve the characteristics of nonlinear CCA and regression while reducing time and space complexity and solving overfitting problems. Cross-sensor images acquired by Landsat-7 Enhanced Thematic Mapper Plus and Landsat-8 Operational and Imager sensors were used to evaluate the proposed method. Experimental results demonstrate the superiority of the proposed method to the related CCA-based methods in terms of computational performance and PIF quality.