查看更多>>摘要：Pansharpening is about fusing a high spatial resolution panchromatic image with a simultaneously acquired multispectral image with lower spatial resolution. In this paper, we propose a Laplacian pyramid pansharpening network architecture for accurately fusing a high spatial resolution panchromatic image and a low spatial resolution multispectral image, aiming at getting a higher spatial resolution multispectral image. The proposed architecture considers three aspects. First, we use the Laplacian pyramid method whose blur kernels are designed according to the sensors' modulation transfer functions to separate the images into multiple scales for fully exploiting the crucial spatial information at different spatial scales. Second, we develop a fusion convolutional neural network (FCNN) for each scale, combining them to form the final multi-scale network architecture. Specifically, we use recursive layers for the FCNN to share parameters across and within pyramid levels, thus significantly reducing the network parameters. Third, a total loss consisting of multiple across scale loss functions is employed for training, yielding higher accuracy. Extensive experimental results based on quantitative and qualitative assessments exploiting benchmarking datasets demonstrate that the proposed architecture outperforms state-of-the-art pansharpening methods. Code is available at https://github.com/ ChengJin-git/LPPN.
查看更多>>摘要：Clustering ensemble integrates multiple base clustering results to obtain a consensus result and thus improves the stability and robustness of the single clustering method. Since it is natural to use a hypergraph to represent the multiple base clustering results, where instances are represented by nodes and base clusters are represented by hyperedges, some hypergraph based clustering ensemble methods are proposed. Conventional hypergraph based methods obtain the final consensus result by partitioning a pre-defined static hypergraph. However, since base clusters may be imperfect due to the unreliability of base clustering methods, the pre-defined hypergraph constructed from the base clusters is also unreliable. Therefore, directly obtaining the final clustering result by partitioning the unreliable hypergraph is inappropriate. To tackle this problem, in this paper, we propose a clustering ensemble method via structured hypergraph learning, i.e., instead of being constructed directly, the hypergraph is dynamically learned from base results, which will be more reliable. Moreover, when dynamically learning the hypergraph, we enforce it to have a clear clustering structure, which will be more appropriate for clustering tasks, and thus we do not need to perform any uncertain postprocessing, such as hypergraph partitioning. Extensive experiments show that, our method not only performs better than the conventional hypergraph based ensemble methods, but also outperforms the state-of-the-art clustering ensemble methods.
查看更多>>摘要：This paper explores a limited trust propagation-based consensus model considering individual attitude for preference modification in a social networked setting with uncertain preference information. To examine the construction of complete linkages, and the status of decision makers in group decision making, it is assumed that the group size and network density will affect the scale of mediators in the propagation process, then a definition of limited trust propagation is proposed and the propagation efficiency can be introduced. On this basis, we obtain missing trust relationships and individual centrality in network. In the process of consensus reaching, both the decision maker's original preference and recommendation advice are considered for flexibly modeling the preference modification process: the individual attitude toward modification is determined by a newly introduced measure of comprehensive relative out-degree centrality, showing the degree of willingness to adjust assessments. When the willingness is too low to reach the preset consensus level, a multi-objective programming model is designed to improve the consensus as much as possible. Moreover, the proposed feedback mechanism narrows the individual acceptable modification range based on the previous adjustment rule, so as to simulate the personalized and targeted decision behavior. To guarantee obtaining a collective aggregated preference in a logical and precise manner, a two-stage optimization model composing of comprehensive relative in-degree centrality-based information aggregation and best consistency -based uncertainty elimination, is proposed. A numerical example and comparative analyses are performed to show the validity and feasibility of the proposed model.
查看更多>>摘要：Accurate retinal vessel segmentation is very challenging. Recently, the deep learning based method has greatly improved performance. However, the non-vascular structures usually harm the performance and some low contrast small vessels are hard to be detected after several down-sampling operations. To solve these problems, we design a deep fusion network (DF-Net) including multiscale fusion, feature fusion and classifier fusion for multi-source vessel image segmentation. The multiscale fusion module allows the network to detect blood vessels with different scales. The feature fusion module fuses deep features with vessel responses extracted from a Frangi filter to obtain a compact yet domain invariant feature representation. The classifier fusion module provides the network more supervision. DF-Net also predicts the parameter of the Frangi filter to avoid manually picking the best parameters. The learned Frangi filter enhances the feature map of the multiscale network and restores the edge information loss caused by down-sampling operations. The proposed end-to-end network is easy to train and the inference time for one image is 41ms on a GPU. The model outperforms state-of-the-art methods and achieves the accuracy of 96.14%, 97.04%, 98.02% from three publicly available fundus image datasets DRIVE, STARE, CHASEDB1, respectively. The code is available at https://github.com/y406539259/DF-Net.
查看更多>>摘要：Multi-view clustering has attracted much attention recently. Among all clustering approaches, spectral ones have gained much popularity thanks to an elaborated and solid theoretical foundation. A major limitation of spectral clustering based methods is that these methods only provide a non-linear projection of the data, to which an additional step of clustering is required. This can degrade the quality of the final clustering due to various factors such as the initialization process or outliers. To overcome these challenges, this paper presents a constrained version of a recent method called Multiview Spectral Clustering via integrating Nonnegative Embedding and Spectral Embedding. Besides retaining the advantages of this method, our proposed model integrates two types of constraints: (i) a consistent smoothness of the nonnegative embedding over all views and (ii) an orthogonality constraint over the columns of the nonnegative embedding. Experimental results on several real datasets show the superiority of the proposed approach.
查看更多>>摘要：Opinion dynamics (OD) models, which simulate individuals' opinion evolution process on social network to analyze the final state of opinion distribution in a group, usually differ from each other due to the differences in social network evolution rules and opinion evolution rules. However, most existing social network evolution rules and opinion evolution rules usually cannot characterize the comprehensive influence of key factors such as neighbors and opinion differences in social relationships. To fully consider the properties of social network evolution and improve the efficiency of consensus reaching process in group decision making, this paper introduces the concept of local world opinion derived from individuals' common friends, and then proposes an individual and local world opinion-based OD model. In the proposed model, social network evolution is jointly determined by the distance between individual opinions and network structure similarity. The pair of individuals with the largest consensus improvement space are then suggested to adjust their opinions by using an adaptive individual opinion adjustment mechanism. Finally, detailed simulation results are provided to demonstrate the convergence of the proposed model and analyze different parameters' effects on the stabilized time steps and the number of stable state opinion clusters.
查看更多>>摘要：Image segmentation is an important issue in many industrial processes, with high potential to enhance the manufacturing process derived from raw material imaging. For example, metal phases contained in microstructures yield information on the physical properties of the steel. Existing prior literature has been devoted to develop specific computer vision techniques able to tackle a single problem involving a particular type of metallographic image. However, the field lacks a comprehensive tutorial on the different types of techniques, methodologies, their generalizations and the algorithms that can be applied in each scenario. This paper aims to fill this gap. First, the typologies of computer vision techniques to perform the segmentation of metallographic images are reviewed and categorized in a taxonomy. Second, the potential utilization of pixel similarity is discussed by introducing novel deep learning-based ensemble techniques that exploit this information. Third, a thorough comparison of the reviewed techniques is carried out in two openly available real-world datasets, one of them being a newly published dataset directly provided by ArcelorMittal, which opens up the discussion on the strengths and weaknesses of each technique and the appropriate application framework for each one. Finally, the open challenges in the topic are discussed, aiming to provide guidance in future research to cover the existing gaps.
NSTL
Elsevier