Sky/cloud images captured by ground-based cameras (a.k.a. whole sky imagers) are increasingly used nowadays because of their applications in a number of fields, including climate modeling, weather prediction, renewable energy generation, and satellite communications. Due to the wide variety of cloud types and lighting conditions in such images, accurate and robust segmentation of clouds is challenging. In this paper, we present a supervised segmentation framework for ground-based sky/cloud images based on a systematic analysis of different color spaces and components, using partial least-squares regression. Unlike other state-of-the-art methods, our proposed approach is entirely learning based and does not require any manually defined parameters. In addition, we release the Singapore whole Sky imaging segmentation database, a large database of annotated sky/cloud images, to the research community.

In recent years, there has been growing interest in using precipitable water vapor (PWV) derived from global positioning system (GPS) signal delays to predict rainfall. However, the occurrence of rainfall is dependent on a myriad of atmospheric parameters. This paper proposes a systematic approach to analyze various parameters that affect precipitation in the atmosphere. Different ground-based weather features such as Temperature, Relative Humidity, Dew Point, Solar Radiation, PWV along with Seasonal and Diurnal variables are identified, and a detailed feature correlation study is presented. While all features play a significant role in rainfall classification, only a few of them, such as PWV, Solar Radiation, Seasonal, and Diurnal features, stand out for rainfall prediction. Based on these findings, an optimum set of features are used in a data-driven machine learning algorithm for rainfall prediction. The experimental evaluation using a 4-year (2012-2015) database shows a true detection rate of 80.4%, a false alarm rate of 20.3%, and an overall accuracy of 79.6%. Compared to the existing literature, our method significantly reduces the false alarm rates.

Near ground radio wave propagation is examined in a tropical plantation experimentally at VHF and UHF bands. The propagation loss with forest depth is empirically analyzed using an integrated model combining both the foliage induced effect and the ground effect. Several well-known empirical foliage models are compared and analyzed. It is observed that the fitted ITU-R model taking into account the ground reflection, can provide a close approximation to the path loss in a tropical palm plantation. However, the accuracy of this model becomes poor when lateral wave dominates in the VHF band. A modification to the ITU-R model is then proposed taking into consideration the lateral wave effect. The modified model is verified using measured and published data, and found to have higher accuracy for large foliage depth in the VHF band as compared to the existing empirical foliage models.

We analyze clouds in the earth's atmosphere using ground-based sky cameras. An accurate segmentation of clouds in the captured sky/cloud image is difficult, owing to the fuzzy boundaries of clouds. Several techniques have been proposed, which use color as the discriminatory feature for cloud detection. In the existing literature, however, analysis of daytime and nighttime images is considered separately, mainly because of differences in image characteristics and applications. In this letter, we propose a lightweight deep-learning architecture called CloudSegNet. It is the first that integrates daytime and nighttime (also known as nychthemeron) image segmentation in a single framework and achieves state-of-the-art results on public databases.

Traditional ground-penetrating radar (GPR) data inversion leverages iterative algorithms that suffer from high computation costs and low accuracy when applied to complex subsurface scenarios. Existing deep learning-based methods focus on the ideal homogeneous subsurface environments and ignore the interference due to clutters and noise in real-world heterogeneous environments. To address these issues, a two-stage deep neural network (DNN), called DMRF-UNet, is proposed to reconstruct the permittivity distributions of subsurface objects from GPR B-scans under heterogeneous soil conditions. In the first stage, a U-shape DNN with first multi-receptive-field convolution (MRF-UNet1) is built to remove the clutters due to inhomogeneity of the heterogeneous soil. Then, the denoised B-scan from MRF-UNet1 is combined with the noisy B-scan to be inputted to the DNN in the second multi-receptive-field convolution (MRF-UNet2). MRF-UNet2 learns the inverse mapping relationship and reconstructs the permittivity distribution of subsurface objects. To avoid information loss, an end-to-end training method combining the loss functions of two stages is introduced. A wide range of subsurface heterogeneous scenarios and B-scans are generated to evaluate the inversion performance. The test results in the numerical experiment and the real measurement show that the proposed network reconstructs the permittivities, shapes, sizes, and locations of subsurface objects with high accuracy. The comparison with existing methods demonstrates the superiority of the proposed methodology for the inversion under heterogeneous soil conditions.

In this paper, a simple algorithm is proposed to perform the nowcasting of rainfall in the tropical region. The algorithm applies global positioning system-derived precipitable water vapor (PWV) values and its second derivative for the short-term prediction of rainfall. The proposed algorithm incorporates the seasonal dependency of PWV values for the prediction of a rain event in the coming 5 min based on the past 30 min of PWV data. This proposed algorithm is based on the statistical study of four-year PWV and rainfall data from a station in Singapore and is validated using two-year independent data for the same station. The results show that the algorithm can achieve an average true detection rate and a false alarm rate of 87.7% and 38.6%, respectively. To analyze the applicability of the proposed algorithm, further validations are done using one-year data from one independent station from Singapore and two-year data from one station from Brazil. It is shown that the proposed algorithm performs well for both the independent stations. For the station from Brazil, the average true detection and false alarm rates are around 84.7% and 37%, respectively. All these observations suggest that the proposed algorithm is reliable and works well with a good detection rate.