Search

Nuclei segmentation, a critical task in cancer diagnosis, presents substantial challenges in histopathological image analysis due to the labor-intensive and time-consuming nature of manual pixel-wise annotation required for fully supervised training. In this study, we address these challenges by proposing a novel weakly supervised segmentation framework. Notably, we introduce two novel methods: color region extraction and sure-foreground extraction. These methods address the limitations of existing approaches by generating high-quality weak masks, effectively mitigating the impact of unwanted image imperfections and irrelevant tissue elements. Furthermore, we enhance the LinkNet architecture by incorporating Dual-branch Spatial Blocks (DBS Blocks) within the decoder section. This novel component empowers the network to capture fine-grained details crucial for accurate segmentation. Finally, we introduce a boundary-aware loss function to refine the model during training, specifically focusing on enhancing nuclei boundary accuracy. Our framework has achieved competitive performance compared to state-of-the-art methods on two benchmark datasets. Notably, our approach effectively addresses unwanted image imperfections and eliminates irrelevant tissue elements, leading to precise segmentation results compared to existing weakly supervised methods. [Display omitted] • Proposing a weakly supervised framework for nuclei segmentation. • Introducing a method for color region extraction in histopathological images. • Innovative method for creating sure-foreground masks. • New component added to LinkNet for better segmentation performance. • Introducing a boundary-aware loss function to refine nuclei boundary precision. [ABSTRACT FROM AUTHOR]

In this paper, we study an event-driven H ∞ control problem of continuous-time nonlinear systems. Initially, with the introduction of a discounted cost function, we convert the nonlinear H ∞ control problem into an event-driven nonlinear two-player zero-sum game. Then, we develop an event-driven Hamilton–Jacobi–Isaacs equation (HJIE) related to the two-player zero-sum game. After that, we propose a novel event-triggering condition guaranteeing Zeno behavior not to happen. The triggering threshold in the newly proposed event-triggering condition can be kept positive without requiring to properly choose the prescribed level of disturbance attenuation. To solve the event-driven HJIE, we employ an adaptive critic architecture which contains a unique critic neural network (NN). The weight parameters used in the critic NN are tuned via the gradient descent method. After that, we carry out stability analysis of the hybrid closed-loop system based on Lyapunov's direct approach. Finally, we provide two nonlinear plants, including the pendulum system, to validate the proposed event-driven H ∞ control scheme. [ABSTRACT FROM AUTHOR]

A brain-controlled vehicle (BCV) is a vehicle that uses the Brain–Computer interface (BCI) to analyze the driver's electroencephalogram (EEG) to obtain human control commands, which can help disabled people extend movement range and improve their self-independence. However, the performance of BCV is poor and the accuracy of BCI recognition varies considerably with different subjects, which severely restrict its promotion and application. To develop a better solution, with the concept of shared control, an intention preferenced fusion brain control strategy based on fuzzy logic is proposed. Firstly, based on Fuzzy Discrete Event System Supervision (FDES) theory, instructions of the driver are supervised and evaluated. At the same time, based on fuzzy logic an automatic controller which gives an automatic decision according to the cur-rent state of the vehicle is designed. Then according to the automatic decision and the correctness of the driver's instructions, a fuzzy reasoning is made, making the final decision more in line with human intentions. In order to verify the effectiveness of the proposed method, based on our previously-designed SSMVEP platform, we conduct virtual experiments, semi-object tests and full physical tests separately for verification. Compared with others shared control schemes, it not only achieves continuous control, but the final control-making is more in line with human intentions, which can encourage users to become more involved, enhancing their self-esteem and brain functions. • An intention preferenced fusion brain control strategy is proposed. • FDES theory is used to supervise the brain instructions. • An automatic controller is designed based on fuzzy logic. • Fuzzy reasoning is used to make the final decision more in line with human intentions. • Virtual experiments, semi-object tests and full physical tests are conducted for verification. [ABSTRACT FROM AUTHOR]

We study the robust stabilization problem of a class of nonlinear systems with asymmetric saturating actuators and mismatched disturbances. Initially, we convert such a robust stabilization problem into a nonlinear-constrained optimal control problem by constructing a discounted cost function for the auxiliary system. Then, for the purpose of solving the nonlinear-constrained optimal control problem, we develop a simultaneous policy iteration (PI) in the reinforcement learning framework. The implementation of the simultaneous PI relies on an actor–critic architecture, which employs actor and critic neural networks (NNs) to separately approximate the control policy and the value function. To determine the actor and critic NNs' weights, we use the approach of weighted residuals together with the typical Monte-Carlo integration technique. Finally, we perform simulations of two nonlinear plants to validate the established theoretical claims. [ABSTRACT FROM AUTHOR]

Gas–liquid two-phase flow is prevalent in both industrial production and the natural environment, making the exploration of its flow behavior critically important. In this study, we design a wire-mesh sensor and measurement system to conduct dynamic experiments on vertical upward gas–liquid two-phase flow, obtaining multi-channel data under various conditions. We develop an integrated limited penetrable visibility graph network (ILPVGNet) to quantitatively assess the dynamic transition from slug flow to bubble flow using network metrics. The results demonstrate that this approach effectively integrates multi-channel measurements, revealing the evolution dynamics and internal coupling mechanisms of gas–liquid two-phase flow. To better observe fine flow structures and details, we perform visualization and super-resolution analysis on the multi-channel data, significantly enhancing image clarity and restoring detailed features, thereby improving our ability to intuitively analyze complex gas–liquid flow behaviors. • A wire mesh sensor is developed for dynamic gas–liquid flow measurement. • ILPVGNet integrates multi-channel data to uncover flow dynamics. • Imaging and super-resolution analysis enhance observation of flow features. [ABSTRACT FROM AUTHOR]

Industry 4.0 is of great significance for the development of oil industry. One of the pivotal steps towards achieving oil Industry 4.0 is accurately mastering the oilfield production dynamics, especially the changes of process parameters. Among many process parameters, the accurate modeling of water cut is extremely critical and difficult. In this paper, under laboratory conditions, oil–water flows with various preset water cuts are simulated. A fluid sensor, equipped with eight concave-shaped conductive electrodes, is employed to capture multi-channel measurement data, continuously recording the oil–water flow process from multiple angles. Subsequently, a novel flow behavior-heuristic deep learning model, named FBHWC model, is developed to model the relationship between the measurement data and water cut, achieving water cut measurement. The FBHWC model is guided by the complex flow behaviors of oil–water flows and consists of two key modules. Particularly, the multi-level feature fusion module focuses on the high-order feature extraction and fusion of sensor measurement data, while the multi-scale measurement module uses fully convolutional design to achieve accurate measurement of water cut. Experimental results show that the FBHWC model has excellent performance in measuring water cut, with mean square error of 0.016%. All these open up a new venue for exploring industrial multiphase flows through combining multi-electrode sensor and deep learning. [ABSTRACT FROM AUTHOR]

An underwater glider (UG) is a buoyancy-propelled and fixed-wing vehicle with attitude controlled completely by means of internal mass redistribution. Since the highly nonlinear and cross-coupled system dynamics, uncertain hydrodynamic parameters and uncertain internal dynamics, model-based control is hardly possible to employ in motion control of UGs. This paper presents a model-free motion control framework of UGs which named inverse model control (IMC). It takes the desired velocity as input, and outputs the control variables of directly and enables the glider to complete the behaviors of motion. Besides, we propose a novel method to represent the framework based on model-free reinforcement learning, which named heterogeneous agent asynchronous policy gradient (HAAPG). It has achieved high-precision motion control and could intermittently adjust the movable mass block rotation amount to correct the deviation caused by currents, which is energy-saving. To verify the effectiveness of the IMC framework and the superiority of HAPPG algorithm, the dynamic model of underwater glider is established and a series of motion simulation scenarios are established. In addition, we conduct sea trials and deploy our IMC framework into actual underwater gliders to perform tasks and achieved satisfactory performance. • A general model-free motion control framework for underwater glider is proposed. • Presented a novel reinforcement learning method to further improve control accuracy. • The proposed framework have completed motion control tasks in simulation results. • Sea trials are conducted to validate the control framework in real world. [ABSTRACT FROM AUTHOR]

The challenge of Fine-grained visual classification (FGVC) comes from the small variations between classes and the large variations within classes. Inspired by the fact that identifying bird species focuses not only on the global features of the subject area but also on the subtle details of the local area, we propose a feature-enhanced Transformer to improve the performance of FGVC. Our proposed method consists of a Dynamic Swin Transformer backbone for extracting comprehensive global image features through continuous attention aggregation, a GCN-based local branch for separating and enhancing local features in different regions, and a pairwise feature interaction (PFI) module for enhancing global features through interactions between image pairs. We conducted extensive experiments on five FGVC datasets to demonstrate the superiority of our method. By fusing the enhanced global and local features, our method achieves the best accuracy compared to existing methods. Our method has an advantage in terms of computational efficiency. • A Feature-Enhanced Transformer for FGVC (FET-FGVC) is proposed. • A Pair-wise Feature Interaction module is designed to enhance global features. • A branch based on GCN is designed to detach and enhance local features. • The model achieves the best performance on four popular FGVC datasets. [ABSTRACT FROM AUTHOR]

Motion state monitoring and recognition are the important issues to be dealt to improve the reliability of Autonomous Underwater Vehicle (AUV). In this work, we transform the motion state classification into Multivariate Time Series Classification (MTSC). By combining two kinds of MTSC methods, including the methods based on feature representation transformation and Deep Neural Network (DNN), we propose a new classification method for Multivariate Time Series (MTS). Multivariate monitoring data of AUV are fused to construct complex networks as graphs to represent the motion states. Then, Graph Convolutional Neural Network (GCNN) is used to extract the features of the graphs and classify the graphs. The effectiveness of our method is validated through sea experiments, whose data are from three classes of navigational motions: near the surface, at fixed depth, and influenced by unknown ocean currents at fixed depth. The experimental results show that the graphical representation based on complex networks can effectively describe the motion states. Compared with Support Vector Machine (SVM), the graphical features are extracted automatically by GCNN to get a higher accuracy of classification of the motion states. The experiments also show that the classification accuracy of our method is higher than that of other two DNNs. • Propose a new method to coarsen multivariate time series as complex network graphs on multi time scales. • Based on graph representation, transform the time series classification into graph classification. • Effectively improve the classification accuracy of motion states of AUV. [ABSTRACT FROM AUTHOR]

• Oblique-incidence reflectivity difference(OIRD) signals characters wax appearance temperature and the rate of change of crystallization process. • OIRD detection reveals the different stages of wax formation process. • OIRD is suitability for in situ probing of wax precipitation of different crude oils. Considering the negative effects of wax precipitation during petroleum transportation on flow assurance and profit loss, in this study, oblique-incidence reflectivity difference (OIRD) analysis is demonstrated to be a powerful tool for the in situ analysis of wax precipitation. Based on the hybrid detection of waxy oil and natural cooling, we develop a detailed wax precipitation model for waxy oil and demonstrate that it can be used to predict actual precipitation features, which are amenable to experimental validation. The curve trend of the OIRD detection exactly reveals the wax formation process, including nucleation, growth, and crystallization. The efficiency of this technique is proven by the fact that the OIRD signals provide the detailed characteristics of the wax precipitation, which include not only the wax appearance temperature (WAT) but also the rate of change of the crystallization process. The advantage of distinguishing between the different stages of the complex wax crystallization process with time and amplitude is demonstrated. [ABSTRACT FROM AUTHOR]

Regional three-dimensional fine oceanic element field reconstruction and feature surface extraction are of considerable significance to oceanographic research and military application. In this paper, a three-dimensional regional oceanic element field reconstruction method based on spatio-temporal Kriging interpolation is proposed, and is studied experimentally. Using discrete data with uneven temporal and spatial distributions, the spatio-temporal variogram of Kriging interpolation is constructed by adopting an integral model. Then, based on experimental data acquired by seven underwater gliders and satellite remote sensing data in the northern South China Sea (NSCS), three-dimensional regional oceanic temperature and salinity fields are constructed for a particular area of interest. Through comparison with the spatial Kriging method and the spatio-temporal weight method, the cross-verification results demonstrate that the proposed method can improve the accuracy of oceanic element field construction. Finally, feature surfaces are extracted from the three-dimensional oceanic element fields, and can be used to reveal thermohaline characteristics of NSCS. Furthermore, the proposed method can be applied in other areas that construct three-dimensional models from discrete data. [ABSTRACT FROM AUTHOR]