Accurate prediction of mesoscale eddy trajectories requires efficient models with large-size available data instances to capture the main eddy characteristics. However, there is a lack of salient attention mechanisms that can recognize the demand of extracting the aggregated features over multidimensional eddy data sources. Additionally, deep learning techniques are very important for eddy trajectory prediction to dynamically capture properties of mesoscale eddies in the South China Sea. In this article, we propose a spatio–temporal attention-based deep learning framework that can orchestrate heterogeneous data integration and propagation trajectory forecast together. It consists of a novel autoencoder equipped with channel and spatial attention mechanisms (CSA-encoder), and a gated recurrent unit (GRU) network with temporal attention layer (TA-GRU). CSA-encoder compresses stereoscopic eddy data with convolutional layers and generates the small-scale and high-quality dataset as the input of TA-GRU. The finer grained TA-GRU method is extended to accurately predict eddy trajectories with more valuable imagery information so that the temporal attention mechanism can automatically select relevant regions within the next 14 days. Our cross-validation results demonstrate that our framework averagely achieves a lower distance error (9 km) and 54% performance improvement over the baseline GRU technique in the next one day, and outperforms two state-of-the-art techniques of long short-term memory and recurrent neural network by 54.9% and 65.6%, respectively.

This work presents a novel framework, named MesoGRU, to accurately predict trajectories of mesoscale eddies (MEs), which is beneficial to study the natural ocean phenomena and perform statistical analysis of oceanic data. MesoGRU first extracts ME trajectory data and establishes two datasets according to the information of location and date in the South China Sea (SCS). Then it effectively processes SCS trajectory data and integrates SLA data and AVISO data into our combined ME dataset (CDME). Furthermore, a designated neural network with a new loss function [(weighted mean square estimation (WMSE)] is designed to learn the characteristics of ME trajectories. After thoroughly analyzing correlations of every two features, the MesoGRU network iteratively extracts key ME features and renormalizes the future time sequence of eddy trajectories. Our verification results demonstrate that MesoGRU has conducted a significant prediction improvement. The mean daily center error of ME trajectory prediction with our scheme is about 8 km and its center error for seven-day forecasting is approximately 18.2 km. Moreover, MesoGRU achieves a higher matching ratio of ME trajectory forecasting compared to other deep learning methods with a single dataset, indicating MesoGRU can predict trajectories as a state-of-the-art method.

This work indicates two novel ocean current observation methods that based on a glider type mobile platform. Glider with current meters are ideal current observation solutions for marine scientists since they have merits of large dynamic area and low power consumption, but their measurement accuracy is insufficient now. The proposed 1st solution, single point current meter compensated inversion method (SPCMCIM), is a revolution of the conventional inverse method by solving the acoustic Doppler to current profiler (AD2CP) induced turbulence problem. It’s a cooperated AD2CP and single point current meter (SPCM) algorithm that compensates the contaminated first AD2CP layer data. The physical position of the SPCM is also optimized according to a finite element analysis (FEA) study to minimize the interference generated by the SPCM. The 2nd method is solving the area restriction problems that AD2CP can just apply to non-pure water environment namely single point and inertial measurement inversion method (SPIMIM). It utilizes the SPCM as the only current observation source and takes advantages of physical properties of ocean currents. The currents are constitute of stable layers and each of them has a stable velocity, so the glider’s velocity variation can be obtained by reading the SPCM. The velocity variation in the transition process between layers can be acquired by the low cost inertial sensors. Finally, these theoretical studies were verified in an onsite sea trial. The SPCM and AD2CP was mounted to an underwater glider to measure the currents and a ship equipped AD2CP was used to provide the arbitrary data. Both of the proposed method can effectively measure the current velocity and induced significant enhancement than the classical inversion method. In particular, the first method conducted improvement of 70% accuracy by average. The 2nd one also shows an obvious improvement and has the advantage to deploy to high purity water area like deep lakes or north pole.

The traditional neural network can maintain the performance of identifying Android malware by learning Android characteristics. But owing to the rapid growth of information on Android devices, it cannot deal with the massive data efficiently for millions of Android applications. In this paper, we propose a RaNetMalDozer framework for malware detection on Android phones to improve the accuracy rate of classification and the training speed. The RaNetMalDozer (RNMD) can dynamically select hidden centers by a heuristic approach and efficiently gather the large-scale datasets of 2550 Android applications by our automatic RNMD data collector. To systematically analyze Android kernel behaviors, we investigate 112 Android kernel features of task_struct and implement EBPN and RNMD methods. Furthermore, compared to the traditional neural network, the EBPN method which achieves 84% accuracy rate, the RNMD method can achieve 94% accuracy rate with the half of training and evaluation time. Our experiments also demonstrate the RNMD method can be used as a better technique of the Android malware detection.

High-resolution synthetic aperture radars (SARs) are becoming an indispensable environmental monitoring system to capture the important geophysical phenomena on the earth and sea surface. However, there is a lack of comprehensive models that can orchestrate such large-scale datasets from numerous satellite missions such as GaoFen-3 and Sentinel-1. In addition, these SAR images of different ocean scenes need to convey a variety of high-level classification features in oceanic and atmospheric phenomena. In this study, we propose a multichannel neural network (MCNN) that supports oceanic SAR scene classification for limited oceanic data samples according to multi-feature fusion, data augmentation, and multichannel feature extraction. To exploit the multichannel semantics of SAR scenes, the multi-feature fusion module effectively combines and reshapes the spatiotemporal SAR images to preserve their structural properties. This fine-grained feature augmentation policy is extended to improve the data quality so that the classification model is less vulnerable to both small- and large-scale data. The multichannel feature extraction also aggregates different oceanic features convolutionally extracted from ocean SAR scenes to improve the classification accuracy of oceanic phenomena with different scales. Through extensive experimental analysis, our MCNN framework has demonstrated a commendable classification performance, achieving an average precision rate of 96%, an average recall rate of 95%, and an average F-score of 95% across ten distinct oceanic phenomena. Notably, it surpasses two state-of-the-art classification techniques, namely, AlexNet and CMwv, by margins of 23.7% and 18.3%, respectively.