Accurate segmentation of very high-resolution remote sensing images is vital for downstream tasks. Most semantic segmentation methods fail to fully consider the inherent characteristics of the images, such as intricate backgrounds, significant intraclass variance, and spatial interdependence of geographic object distribution. To address these challenges, we propose an efficient global–local scene awareness network with rotary position embedding (RoGLSNet). Specifically, we introduce the dynamic global filter (DGF) module to adaptively select frequency components, thereby mitigating interference from background noise. For high intraclass variance, the class center aware block (CCAB) performs class-level contextual modeling with spatial information integration. Additionally, the rotary position embedding (RoPE) is incorporated into vanilla attention to indirectly model the positional and distance relationships of geographic target objects. Extensive experimental results on two widely used datasets demonstrate that RoGLSNet outperforms the state-of-the-art (SOTA) segmentation methods. The code is available athttps://github.com/bai101315/RoGLSNet

High-speed trains are susceptible to unexpected events such as strong winds and equipment failures, which can result in deviations from the scheduled timetable. As the density of traffic increases, these delays can quickly spread to other trains, eventually leading to conflicts in the timetable. To ensure the efficiency of high-speed railways, quickly resolving potential conflicts and generating appropriate rescheduling schemes are essential. The existing hierarchical structure of train control and online rescheduling tends to be inefficient in terms of information communication and can even lead to unfeasible rescheduled timetables and trajectories. To address these issues, an integrated structure of timetable rescheduling and train trajectory optimization is proposed by introducing the train minimum running time into the process of timetable rescheduling and using the adjusted running time as the objective of trajectory optimization. The integration model is formulated by considering the constraints of timetable rescheduling such as the maximum number of trains overtaking trains, platforms at stations, and the priority of the train, as well as the constraints of trajectory optimization. A deep reinforcement learning (DRL)-based approach is proposed to solve the problem. Numerical experiments are conducted on a segment of the Beijing-Shanghai high-speed railway line, using adapted data to demonstrate the effectiveness of the proposed method in rescheduling timetables and optimizing train trajectories. The results show that the integrated rescheduled timetable and the optimized train trajectory can be generated simultaneously and the computation time exhibits a linear increase with respect to the size of the problem.

While text-based captchas, designed to distinguish between human users and bots, have encountered numerous attack methods, they remain a prevalent security mechanism employed by various websites. Some deep learning-based approaches can recognize captcha character sequences end-to-end; however, the labor-intensive and time-consuming labeling process severely restricts their feasibility. In this study, we introduce CaptchaSAM, to segment anything in text-based captchas. Our insight lies in the fact that identifying individual characters is a simpler task compared to recognizing character sequences, leading to a substantial reduction in labeling dependency. To accomplish this, we utilize the Segment Anything Model (SAM) for character-level semi-automatic annotation. Subsequently, we leverage the annotated data to train a semantic segmentation model. Our experiments with real-world captcha systems demonstrate that CaptchaSAM significantly outperforms state-of-the-art methods with just a few labeled captchas. We anticipate that our research will encourage security experts to reconsider the design and deployment of text-based captchas. The source code is accessible at https://github.com/SJTU-dxw/CaptchaSAM.

This paper addresses the cooperative control problem of trains with specific consideration of human interventions in unusual situations where hard handovers on control objectives and safety constraints may occur. To overcome the effects of such discontinuous factors caused by interventions on the smooth operation of trains and avoid drastic changes in the control input, cooperative control policies with soft-switching are constructed based on novel potential energy functions and weighted functions such that, besides achieving consensus among trains for desired velocities and positions, maintaining prescribed tracking distance, collision avoidance is also guaranteed during the state transition process. Furthermore, adaptive approximation and saturation compensation mechanisms are adopted to cope with parameter uncertainties and input saturation. A rigorous proof is provided to demonstrate the correctness of the proposed results theoretically, and numerical experiments are conducted using real operation data to illustrate the theoretical conclusions.

Glaucoma is an eye disease that leads to irreversible vision loss. Accurate Optic Disc (OD) and Optic Cup (OC) segmentation can effectively facilitate ophthalmologist in glaucoma diagnosis. Recently, a series of deep learning approaches attain promising performance in OD and OC segmentation but still face the challenge to precisely segment OC boundary with enhanced computational efficiency. To address this issue, we propose a novel network named Robust Multiscale Feature Extraction with Depthwise Separable Convolution (RMSDSC-Net), which can better solve the challenging tradeoff between segmentation performance and network cost. The proposed RMSDSC-Net is mainly composed of Multiscale Input (MSI), Depthwise Separable Convolution Unit (DSCU), Dilated Convolution Block (DCB), and External Residual Connection (ERC). First, the introduction of MSI can reduce the information loss due to the pooling layers used in the network for capturing rich feature representations. Next, to enhance segmentation performance and computational efficiency, this paper designs DSCU and DCB modules to avoid spatial information loss from minor details of the image and preserve more high-level semantic features. Finally, this paper develops ERC established between the encoding layers and decoding layers to minimize the feature degradation problem. Hence, a high segmentation performance can be achieved using a shallow network. To evaluate the performance of the proposed network, extensive experiments have been enforced on two publicly available databases, DRISHTI-GS and REFUGE. Our approach outperforms the state-of-the-art approaches with the Dice Coefficient of (0.978, 0.919) and (0.965, 0.910) for OD and OC segmentation on DRISHTI-GS and REFUGE databases, respectively. As a result, the proposed approach has a strong potential in analyzing fundus images for glaucoma diagnosis.

The coordinated control problem for a multiple high-speed train (HST) system subject to unknown communication delays is systematically investigated in this paper. Taking into consideration the inertial lag of the servo motor, a third-order nonlinear control model is constructed to capture the dynamics of a train in real-world operations. By virtue of the backstepping linearization technique, the coordinated control of trains is formulated as a stabilization problem for a linear multiple-input multiple-output system with an unknown input delay. Distributed control laws with a time-varying low gain parameter are designed, besides solving the stabilization problem, to guarantee a fast convergency rate during the train status adjustment process. Numerical examples are provided to illustrate that the time-varying low gain parameter design achieves better control performance compared with the traditional constant low gain feedback design in terms of the convergency rate and the system overshot, and that the proposed control method is effective in train tracking distance adjustment.

The cruise control problems for high-speed trains are investigated in this paper. Both a single train and multiple trains on a railway line are considered. The cars in a single train are modeled as a group of ordered particles connected by flexible couplers. Each car is viewed as an intelligent agent that communicates with its neighbors, making the train a multi-agent system. The information transmission topology among these agents is represented by a connected undirected graph. Distributed cooperative control laws are constructed that achieve displacement and speed consensus among cars at a desired profile, while guaranteeing the coupler displacements to be within a safety range and converge to the nominal value. For multiple trains on a railway line, each train has access to the information of the trains within its wireless communication range, making all cars in these trains a multi-agent system. The underlying communication topology is now a state-dependent undirected graph. Distributed control laws are designed such that, besides achieving coordinated control of cars among each train, consensus among trains at the desired displacement and speed profile and connectivity among trains are also achieved, while avoiding collision. Extensive simulation results are presented to illustrate the theoretical conclusions we have reached.