Accurate and robust detection of multi-class objects in very high resolution (VHR) aerial images has been playing a significant role in many real-world applications. The traditional detection methods have made remarkable progresses with horizontal bounding boxes (HBBs) due to CNNs. However, HBB detection methods still exhibit limitations including the missed detection and the redundant detection regions, especially for densely-distributed and strip-like objects. Besides, large scale variations and diverse background also bring in many challenges. Aiming to address these problems, an effective region-based object detection framework named Multi-scale Feature Integration Attention Rotation Network (MFIAR-Net) is proposed for aerial images with oriented bounding boxes (OBBs), which promotes the integration of the inherent multi-scale pyramid features to generate a discriminative feature map. Meanwhile, the double-path feature attention network supervised by the mask information of ground truth is introduced to guide the network to focus on object regions and suppress the irrelevant noise. To boost the rotation regression and classification performance, we present a robust Rotation Detection Network, which can generate efficient OBB representation. Extensive experiments and comprehensive evaluations on two publicly available datasets demonstrate the effectiveness of the proposed framework.

Over-the-horizon radar (OTHR) target tracking in the presence of complicated ionospheric environment mainly faces three challenges, i.e., discrete uncertainty of multipath data association, continuous uncertainty of ionospheric heights, and coupling of target state estimation and ionospheric parameters identification. The existing OTHR target tracking algorithms demanded that the ionospheric heights should be exactly known or statistical properties known. However, the ionospheric heights is inaccurate due to the inherent variability of ionosphere, especially when the deployment of ionosondes is unavailable in the sea area or hostile zone. This paper introduces a joint optimization scheme called distributed expectation-conditional maximization (DECM), which solves the target state estimation, multipath data association, and ionospheric heights identification simultaneously. The proposed DECM algorithm consists of a local estimation level and a global fusion level whereas information is exchanged within these two levels until iteration terminates. This dual-level processing framework transforms the high-dimensional estimation problems into several low-dimensional parallel path-dependent estimation problems, which improves the computational efficiency of expectation maximization under high-dimensional latent variables case. In addition, the closed-loop structure is beneficial to the coupling problem. The simulation indicates the effectiveness of the proposed scheme.

This paper addresses the challenges posed by imperfect detection and uncertain parameters, such as detection probability and noise covariances, in target tracking. We introduce an Adaptive Bernoulli Filter (ABF) capable of handling multiple sources of uncertainty simultaneously. The ABF employs a Gaussian Inverse Gamma Inverse Wishart Mixture (GIGIWM) to represent the spatial probability density function of the augmented state. Using a variational Bayesian approach, we derive a closed-form solution for the filter, providing estimates for target existence probability, kinematic and feature states, measurement noise covariance matrix, and predicted error covariance matrix. Additionally, we extend the ABF to incorporate prior knowledge through constrained distributions. In a distributed multi-sensor scenario, we propose a fusion approach to combine local posteriors, extending existing fusion techniques to handle local posteriors that depend on both global and local variables. Simulation results show the effectiveness and robustness of the proposed filter and distributed fusion framework.

Intent inference has attracted considerable interest for achieving situation awareness in the high-level information fusion community. Different from traditional tracking-then-inference methods for intent inference, a novel scheme for joint target tracking, classification and intent inference (JTCI) is developed based on the Bayesian framework. The proposed JTCI scheme exploits the dependence of target state on target class and intent by defining intent and class dependent dynamic model sets. Then, the joint target state, intent and class density is obtained recursively under the assumption that the kinematic and attribute measurement processes are conditional independent. Finally, simulations about tracking in the air surveillance system are utilized to demonstrate the superiority of the proposed JTCI to the state-of-the-art JTC.

This paper concentrates on a globally sequential fusion state estimation problem for clustered wireless sensor networks (WSNs). Therein, frequent data communications in WSN will significantly increase energy consumption and communication burden, and it needs to be taken into account. To reduce unnecessary data transmissions and save energy, an event-triggered mechanism that decides whether the current measurement should be transmitted or not, therefore, is introduced. Then a novel variational Bayesian based event-triggered sequential measurement fusion (VB-ESMF) estimator is proposed to produce the local fused results of the clustered WSNs, where the variational Bayesian approach is used to infer the measurement noise covariance matrices of pseudo measurement noises. Therein, the local fused measurements of the clusters in the clustered WSNs are used sequentially by the remote fusion center, and thus a global fusion state estimation, which is globally optimal when all measurements are transmitted successfully, is computed. Additionally, certain boundedness and convergence conditions of the proposed estimator are derived, and an expected compromise between communication rate and estimation accuracy can be obtained by properly turning the trigger threshold. Finally, the effectiveness of both the VB-ESMF estimation and the globally sequential fusion state estimation is illustrated by simulation results.