17 April 2024 in IEEE Sensors Journal
In recent years, the use of Mobile Charging Vehicle (MCV) to simultaneously replenish energy for multiple nodes has become a research focus in Wireless Rechargeable Sensor Networks (WRSNs). The selection of charging thresholds is crucial to enhance the charging efficiency of WRSNs. However, most of the existing work use a fixed threshold, which has to face the risk of frequent charging triggers when the threshold is too high, or data loss when the threshold is too low. Furthermore, this challenge will become even more severe in the fact that the charging queue changes. Therefore, this paper first proposes an Adaptive Dual-Threshold Online Selection Algorithm (ADT-OSA) based on the distance between nodes, energy consumption rate, and remaining energy to determine upper and lower charging thresholds for each node. Subsequently, an Improved Real-time Genetic Algorithm (IR-IGA) is introduced based on the overall and real-time energy situation of WRSNs to dynamically determine the charging path for MCV. Finally, the performance of the two algorithms is evaluated through extensive simulations. The results show that the adaptive double threshold by the ADT-OSA algorithm is better than the double threshold obtained by trial and error method, and the IR-IGA algorithm significantly reduces the mobility energy consumption, charging energy consumption and the amount of data loss compared with existing methods.
28 September 2022 in arXiv
Space-based infrared tiny ship detection aims at separating tiny ships from the images captured by earth orbiting satellites. Due to the extremely large image coverage area (e.g., thousands square kilometers), candidate targets in these images are much smaller, dimer, more changeable than those targets observed by aerial-based and land-based imaging devices. Existing short imaging distance-based infrared datasets and target detection methods cannot be well adopted to the space-based surveillance task. To address these problems, we develop a space-based infrared tiny ship detection dataset (namely, NUDT-SIRST-Sea) with 48 space-based infrared images and 17598 pixel-level tiny ship annotations. Each image covers about 10000 square kilometers of area with 10000X10000 pixels. Considering the extreme characteristics (e.g., small, dim, changeable) of those tiny ships in such challenging scenes, we propose a multi-level TransUNet (MTU-Net) in this paper. Specifically, we design a Vision Transformer (ViT) Convolutional Neural Network (CNN) hybrid encoder to extract multi-level features. Local feature maps are first extracted by several convolution layers and then fed into the multi-level feature extraction module (MVTM) to capture long-distance dependency. We further propose a copy-rotate-resize-paste (CRRP) data augmentation approach to accelerate the training phase, which effectively alleviates the issue of sample imbalance between targets and background. Besides, we design a FocalIoU loss to achieve both target localization and shape description. Experimental results on the NUDT-SIRST-Sea dataset show that our MTU-Net outperforms traditional and existing deep learning based SIRST methods in terms of probability of detection, false alarm rate and intersection over union.