27 October 2023 in Signal Processing
2 September 2024 in IEEE Signal Processing Letters
In this letter, a new reconstructed variational Bayesian (RVB) Kalman filter is proposed for handling heavy-tailed and skewed noises. The Student's t distribution is used to model predicted state, and the skewed Gaussian-exponential mixture distribution is used to model measurement noises. In order to more accurately deal with process noises, we model the predicted covariance based on the inverse Wishart distribution. Subsequently, a new RVB framework is proposed to separate the iterations of process and measurement noise related parameters, which achieves decoupling of hidden variables. Under the new RVB framework, the RVB-ST-SGEM-KF algorithm is derived. The superiority of the proposed RVB framework and algorithm are validated through a maneuvering target tracking scenario.
26 September 2023 in Measurement Science and Technology
Considering that existing robust filtering algorithms rely on the selection of initial values of degree of freedom (DOF) parameters in outlier interference environments and cannot effectively cope with unknown non-stationary heavy-tailed measurement noises (HMN), a Gaussian–Pearson type VII (PTV) adaptive mixture distribution-based outlier-robust Kalman filter (GPTVMAKF) is proposed. In order to determine whether the current measurement is a normal value or an outlier, a judgment factor subject to the Beta-Bernoulli distribution is introduced. PTV distribution is used to model HMN caused by outliers, and two Gamma distributions are used to model the two different DOF parameters, which can make the PTV distribution have the adaptive adjustment ability. By introducing the inverse Wishart distribution as the prior distribution of the measurement noise covariance, which is adaptively estimated to cope with the unknown time-varying measurement noises. The state and parameters are jointly estimated by variational Bayesian. Finally, the simulation experiments verify that the proposed GPTVMAKF can obtain more accurate state estimation than existing filters in the environments with varying degrees of HMN and unknown non-stationary HMN.
27 August 2025 in IEEE Transactions on Aerospace and Electronic Systems
The main challenges faced in detecting infrared small targets include small target sizes, lack of clear shape and texture information, and low signal-to-noise ratio in complex environments. We propose an infrared small target detection network, FSFANet, to overcome these difficulties. The network improves the representation of features for small targets through the integration of a prior knowledgebased attention module (PAM) and a multi-scale dilated convolution module (MDCM). Additionally, the U-Net architecture is improved to effectively integrate low-level details and high-level semantic information through a full-scale skip connection strategy. To further incorporate features of different scales, the axial self-cross attention module (ASCA) is introduced. This module can capture long-range dependencies and encode multi-scale information while maintaining computational efficiency. In addition, this paper introduces a hybrid loss function that combines weighted cross-entropy loss and SoftIoU loss to effectively address the issue of imbalance between target and background pixels, thereby enhancing the segmentation accuracy of the model. Experimental results on multiple datasets validate the outstanding performance of FSFANet with a relatively small number of parameters, demonstrating its effectiveness in applying small infrared target detection.