The latest remote sensing image saliency detectors primarily rely on RGB information alone. However, spatial and geometric information embedded in depth images is robust to variations in lighting and color. Integrating depth information with RGB images can enhance the spatial structure of objects. In light of this, we innovatively propose a remote sensing image saliency detection model that fuses RGB and depth information, named the multimodal guided transformer architecture (MGTA). Specifically, we first introduce the strong correlated complementary fusion (SCCF) module to explore cross-modal consistency and similarity, maintaining consistency across different modalities while uncovering multidimensional common information. Additionally, the global-local context information interaction (GLCII) module is designed to extract global semantic information and local detail information, effectively utilizing contextual information while reducing the number of parameters. Finally, a cascaded feature-guided decoder (CFGD) is employed to gradually fuse hierarchical decoding features, effectively integrating multi-level data and accurately locating target positions. Extensive experiments demonstrate that our proposed model outperforms 14 state-of-the-art methods. The code and results of our method are available at https://github.com/Zackisliuzao/MGTANet.

Moving object detection (MOD) in satellite video sequences faces persistent challenges including low contrast against complex backgrounds, limited motion modeling, and severe scale variation. To effectively address these, we propose STFDNet, a novel combined model-driven and data-driven framework. STFDNet comprises three core components: the hybrid temporal motion module (HTMM), dynamic frequency learning (DFL), and a progressive cascaded learning strategy (PCLS). Initially, HTMM leverages both explicit and implicit strategies to model multi-frame temporal differences, effectively capturing both short- and long-term motion. Then, DFL integrates a frequency-domain dynamic filter to learn global frequency characteristics, enhancing feature representation beyond the spatial domain and improving distinction from background noise. Finally, PCLS progressively refines detection results by fusing multi-domain features in a step-by-step manner, transitioning from coarse-grained to fine-grained representations. Experiments on the Jilin-1 satellite video dataset demonstrate that the proposed method effectively mitigates background noise, limited motion modeling, and object scale variations, significantly improving detection accuracy and robustness, achieving an F1-score of 85.4%.

In recent years, salient object detection (SOD) has attracted more and more attention. However, the SOD in remote sensing images (RSI-SOD) faces various issues, including large scene span, cluttered background and changeable object scale. To address these challenges, an edge complementary multi-scale aggregation network (ECMANet) is proposed in this paper. Specifically, a multi-scale feature aggregation module (MFAM) is designed to extract hierarchical multi-scale information and reduce the noise interference of different scale information. In addition, foreground edge guidance module (FEGM) is designed to cross-refine foreground information and edge information. Finally, the foreground, edge, and background are generated by background-foreground fusion module (BFFM) to complement the overall network information. Extensive experiments are conducted on two popular datasets demonstrate that the proposed method outperforms other state-of-the-art methods.

In recent years, numerous outstanding technologies have been proposed for salient object detection (SOD) in remote sensing images (RSIs), but most of them focus solely on improving performance while disregarding computational, thereby lacking portability and mobility. This article introduces a novel lightweight progressive multilevel feature collaborative network, termed LPMFCNet. This framework constructs progressive feature information through multilevel image content extraction and designs a multichannel interactive deep neural network with information fusion and filtering functions. First, a spatial detail enhancement module (SDEM) is devised to acquire distant feature information through intermediate branch expansion of receptive fields while preserving multiscale information extraction. Second, an advanced semantic interaction module (ASIM) is proposed to model distant dependency relationships between deep semantic features to better identify the positional information of salient objects. Finally, a multilevel feature collaboration module (MFCM) is designed to collaboratively utilize target features from a multilevel perspective, which fully mining deep-level semantic positional information while retaining target detail information. Extensive experimental comparisons are conducted on two remote sensing datasets with 17 advanced methods. Results demonstrate that the proposed method exhibits superior detection performance while maintaining lightweightness. The LPMFCNet only contains 3.26M parameters and runs 0.5G FLOPs for a$256\times 256$image.

Convolutional neural network (CNN) model has made a great breakthrough in target detection in remote sensing image due to the excellent feature extraction capability. However, diverse scenes and complex contextual information of remote sensing image make these CNN models face big challenges. For example, the distinctiveness between the target and the context would be reduced greatly. This letter proposes an object-and-scene context constrained CNN method to detect target in remote sensing image. This method has two channels, namely, object context constrained channel and scene context constrained channel. The object context constrained channel uses recurrent neural network (RNN) to explore the contextual relationship between the target and the object, including feature relationship and position relationship. The scene context constrained channel adopts priori scene information and Bayesian criterion to infer the relationship between the scene and the target, and it make full use of the scene information to enhance the target detection performance. The experimental results on two datasets demonstrate the robustness and effectiveness of the proposed method.

Remote sensing image has been widely used in many fields such as military reconnaissance and earthquake relief. However, the complexity and diversity of the scene make the target detection and recognition performance poor. Convolutional neural networks (CNNs) have made breakthrough in remote sensing image processing due to their ability to extract deep features. This letter proposes a remote sensing image scene recognition method based on object relationship reasoning CNN (ORRCNN), which makes use of the relationship between objects to infer the scene information. The method has prior scene-information-based channel and object-detection-based channel to classify the remote sensing image. The prior scene-information-based channel makes use of the feature space to identify the scene, and the object-detection-based channel adopts the relationship between the object and the scene to classify the scene. Afterward, the Bayesian criterion infers the scene more accurately by means of fusing the scene information from the above channels. The experimental results show that the proposed method is excellent especially in the scene where there are iconic objects in the remote image.

Cyber-Physical Systems (CPSs), inevitably exposed in open environment, are considered to be complex to analyze in terms of its intrinsic heterogeneity and potential stochastic behaviors. Some existing technologies like Functional Mock-up Interface (FMI) could mitigate the issue to some extent, however there are still some challenging problems, for example, effective and efficient co-simulation for stochastic models like markov chains. To facilitate co-simulation of stochastic CPSs, we present an improved co-simulation framework that focuses on the capture of nearest future event to reduce the number of running steps and the frequency of data exchange between models. The core implementation is an adaptive co-simulator that integrates two algorithms optimized with event detection for co-simulation between Functional Mock-up Unit (FMU) and two different types of markov chains: DTMC and CTMC. Meanwhile, a Prism wrapper is implemented for interpreting a markov chain as a fake FMU. To demonstrate the ability of our improved co-simulation, we study two extended Bouncing Ball cases respectively modelled by DTMC and CTMC. The experiment result turns out that our approach is effective in generating simulation traces of stochastic CPSs and the optimized algorithms are more efficient compared with original one.

Cyber-Physical Systems (CPSs) as advanced embedded systems integrating computation with physical process are increasingly penetrating into our life. Modeling and analysis for such systems closely involved with us are actively researched. A current challenging problem is how to take advantages of existing technologies like SysML/MARTE, Modelica and Statistical Model Checking (SMC) through effective integration. Moreover, the lack of efficient methodologies or tools for modeling and analysis of CPSs makes the gap between design and analysis models hard to bridge. To solve these problems, we present a framework named Modana to achieve an integrated process from modeling with SysML/MARTE to analysis with SMC for CPSs in terms of Non Functional Properties (NFP) such as time, energy, etc. Functional Mock-up Interface (FMI), as a connecting link between modeling and analysis, plays a major role in coordinating various tools for co-simulation to generate traces as the input of statistical model checker. To demonstrate the capability of Modana framework, we model energy-aware buildings as a case study, and discuss the analysis on energy consumption in different scenarios.

Cyber-Physical Systems (CPSs) face many challenges, one of which is the complexity of our world full of a variety of stochastic behavior. Due to the excess complexity the increasing number of need for autonomous long running components appears and gives rise to a special concern for energy so that a great challenge becomes open to us that how to model, analyze and make effective evaluation for either one or both of stochastic behavior and energy consumption. To solve the problem, we present a Stochastic Occurrence Hybrid Automata (SOHA) which unify all stochastic behavior into triggers among probabilistic events and use a unified way to describe both stochastic and deterministic events occurrence, besides introduce the energy function with time to model energy harvesting or consumption. In this paper, we give the formal syntax and semantics of SOHA based on labeled transition system and then propose a SOHA-based modeling approach that provides a more reasonable way to concisely model stochastic hybrid systems with the use of refinement and stochastic abstraction. This approach helps build a better model with hiding the details we may not concern, which is useful to the analysis in the future. To illustrate our approach and its benefit, we discuss a benchmark of hybrid systems Energy Aware Buildings as case study.