Multi-resolution Network Research Articles

An Image Steganalysis Algorithm Based on Multi-Resolution Feature Fusion

The misuse of image steganography poses significant risks to societal security. Whether images include concealed data is a critical problem of information security. Traditional convolutional layers often fail to adequately capture the global correlation of steganographic features as network depth increases, leading to redundant model parameters and missing key features, thereby weakening steganographic signal detection. To solve the problems of highly covert steganographic algorithms and the weakness of traditional methods, a steganography detection solution using multi-resolution feature fusion is presented. This approach uses a multi-resolution network to increase the interactivity from higher to lower resolution. The results of the experiments confirm that the proposed algorithm allows for a maximum accuracy of 90.56% for the embedding rate of 0.4bpp. The overall results prove that the proposed model achieves higher accuracy and better performance than some leading steganalysis models available when applied to different steganographic algorithms and embedding rate conditions.

HRNet: Differentially Private Hierarchical and Multi-Resolution Network for Human Mobility Data Synthesization

Human mobility data offers valuable insights for many applications such as urban planning and pandemic response, but its use also raises privacy concerns. In this paper, we introduce the Hierarchical and Multi-Resolution Network (HRNet), a novel deep generative model specifically designed to synthesize realistic human mobility data while guaranteeing differential privacy. We first identify the key difficulties inherent in learning human mobility data under differential privacy. In response to these challenges, HRNet integrates three components: a hierarchical location encoding mechanism, multi-task learning across multiple resolutions, and private pre-training. These elements collectively enhance the model's ability under the constraints of differential privacy. Through extensive comparative experiments utilizing a real-world dataset, HRNet demonstrates a marked improvement over existing methods in balancing the utility-privacy trade-off.

Agent-based modelling of polarized news and opinion dynamics in social networks: a guidance-oriented approach

Abstract In the contemporary social network landscape, opinion polarization has emerged as a prominent challenge, sparking concerns about the effective guidance of news sentiment and mitigation of opposing opinions. This is particularly pertinent in the intricate web of social networks, where complexity reigns supreme. Addressing this pivotal issue, this article introduces a news opinion guidance approach grounded in motif recognition. To accurately mirror real-world social networks, we have crafted an agent-based model that simulates polarized news propagation. This model encompasses diverse media agents and user agents, meticulously replicating the news dissemination process within the network. In our quest to unveil the underlying structures of social networks, we have enhanced the Augmented Multiresolution Network approach, incorporating multi-dimensional node attributes for more nuanced clustering and network mapping. This refinement enables us to pinpoint potential motif regions with greater precision. Leveraging these insights, we introduce a triangular motif-based opinion guidance strategy aimed at shaping opinion distribution by bolstering the influence of nodes within these motifs. Once the pertinent motifs are identified, we undertake simulation experiments that reveal the remarkable efficacy of our motif recognition-driven guidance strategy. Notably, it reduces opinion polarization by a substantial 74% compared to scenarios without guidance strategies. This research offers a fresh perspective on crafting personalized and targeted news sentiment guidance strategies. It presents a versatile and potent computational framework for understanding and managing polarization phenomena in social networks, carrying profound theoretical and practical ramifications.

Image inpainting via multi-resolution network with Fourier convolutions

Image inpainting via multi-resolution network with Fourier convolutions

A novel multi-resolution network for the open-circuit faults diagnosis of automatic ramming drive system

The open-circuit fault is one of the most common faults of the automatic ramming drive system (ARDS), and it can be categorized into the open-phase faults of Permanent Magnet Synchronous Motor (PMSM) and the open-circuit faults of Voltage Source Inverter (VSI). The stator current serves as a common indicator for detecting open-circuit faults. Due to the identical changes of the stator current between the open-phase faults in the PMSM and failures of double switches within the same leg of the VSI, this paper utilizes the zero-sequence voltage component as an additional diagnostic criterion to differentiate them. Considering the variable conditions and substantial noise of the ARDS, a novel Multi-resolution Network (MrNet) is proposed, which can extract multi-resolution perceptual information and enhance robustness to the noise. Meanwhile, a feature weighted layer is introduced to allocate higher weights to characteristics situated near the feature frequency. Both simulation and experiment results validate that the proposed fault diagnosis method can diagnose 25 types of open-circuit faults and achieve more than 98.28% diagnostic accuracy. In addition, the experiment results also demonstrate that MrNet has the capability of diagnosing the fault types accurately under the interference of noise signals (Laplace noise and Gaussian noise).

Multi-resolution network based image steganalysis model

Multi-resolution network based image steganalysis model

High-Resolution Network with Dynamic Convolution and Coordinate Attention for Classification of Chest X-ray Images.

The development of automatic chest X-ray (CXR) disease classification algorithms is significant for diagnosing thoracic diseases. Owing to the characteristics of lesions in CXR images, including high similarity in appearance of the disease, varied sizes, and different occurrence locations, most existing convolutional neural network-based methods have insufficient feature extraction for thoracic lesions and struggle to adapt to changes in lesion size and location. To address these issues, this study proposes a high-resolution classification network with dynamic convolution and coordinate attention (HRCC-Net). In the method, this study suggests a parallel multi-resolution network in which a high-resolution branch acquires essential detailed features of the lesion and multi-resolution feature swapping and fusion to obtain multiple receptive fields to extract complicated disease features adequately. Furthermore, this study proposes dynamic convolution to enhance the network's ability to represent multi-scale information to accommodate lesions of diverse scales. In addition, this study introduces a coordinate attention mechanism, which enables automatic focus on pathologically relevant regions and capturing the variations in lesion location. The proposed method is evaluated on ChestX-ray14 and CheXpert datasets. The average AUC (area under ROC curve) values reach 0.845 and 0.913, respectively, indicating this method's advantages compared with the currently available methods. Meanwhile, with its specificity and sensitivity to measure the performance of medical diagnostic systems, the network can improve diagnostic efficiency while reducing the rate of misdiagnosis. The proposed algorithm has great potential for thoracic disease diagnosis and treatment.

β-catenin/CBP activation of mTORC1 signaling promotes partial epithelial-mesenchymal states in head and neck cancer

Head and neck cancers, which include oral squamous cell carcinoma (OSCC) as a major subsite, exhibit cellular plasticity that includes features of an epithelial-mesenchymal transition (EMT), referred to as partial-EMT (p-EMT). To identify molecular mechanisms contributing to OSCC plasticity, we performed a multiphase analysis of single cell RNA sequencing (scRNAseq) data from human OSCC. This included a multiresolution characterization of cancer cell subgroups to identify pathways and cell states that are heterogeneously represented, followed by casual inference analysis to elucidate activating and inhibitory relationships between these pathways and cell states. This approach revealed signaling networks associated with hierarchical cell state transitions, which notably included an association between β-catenin-driven CREB-binding protein (CBP) activity and mTORC1 signaling. This network was associated with subpopulations of cancer cells that were enriched for markers of the p-EMT state and poor patient survival. Functional analyses revealed that β-catenin/CBP induced mTORC1 activity in part through the transcriptional regulation of a raptor-interacting protein, chaperonin containing TCP1 subunit 5 (CCT5). Inhibition of β-catenin-CBP activity through the use of the orally active small molecule, E7386, reduced the expression of CCT5 and mTORC1 activity in vitro, and inhibited p-EMT-associated markers and tumor development in a murine model of OSCC. Our study highlights the use of multiresolution network analyses of scRNAseq data to identify targetable signals for therapeutic benefit, thus defining an underappreciated association between β-catenin/CBP and mTORC1 signaling in head and neck cancer plasticity.

Improved distinct bone segmentation in upper-body CT through multi-resolution networks

PurposeAutomated distinct bone segmentation from CT scans is widely used in planning and navigation workflows. U-Net variants are known to provide excellent results in supervised semantic segmentation. However, in distinct bone segmentation from upper-body CTs a large field of view and a computationally taxing 3D architecture are required. This leads to low-resolution results lacking detail or localisation errors due to missing spatial context when using high-resolution inputs.MethodsWe propose to solve this problem by using end-to-end trainable segmentation networks that combine several 3D U-Nets working at different resolutions. Our approach, which extends and generalizes HookNet and MRN, captures spatial information at a lower resolution and skips the encoded information to the target network, which operates on smaller high-resolution inputs. We evaluated our proposed architecture against single-resolution networks and performed an ablation study on information concatenation and the number of context networks.ResultsOur proposed best network achieves a median DSC of 0.86 taken over all 125 segmented bone classes and reduces the confusion among similar-looking bones in different locations. These results outperform our previously published 3D U-Net baseline results on the task and distinct bone segmentation results reported by other groups.ConclusionThe presented multi-resolution 3D U-Nets address current shortcomings in bone segmentation from upper-body CT scans by allowing for capturing a larger field of view while avoiding the cubic growth of the input pixels and intermediate computations that quickly outgrow the computational capacities in 3D. The approach thus improves the accuracy and efficiency of distinct bone segmentation from upper-body CT.

Self-Parameterization Based Multi-Resolution Mesh Convolution Networks

This paper addresses the challenges of designing mesh convolution neural networks for 3D mesh dense prediction. While deep learning has achieved remarkable success in image dense prediction tasks, directly applying or extending these methods to irregular graph data, such as 3D surface meshes, is nontrivial due to the non-uniform element distribution and irregular connectivity in surface meshes which make it difficult to adapt downsampling, upsampling, and convolution operations. In addition, commonly used multiresolution networks require repeated high-to-low and then low-to-high processes to boost the performance of recovering rich, high-resolution representations. To address these challenges, this paper proposes a self-parameterization-based multi-resolution convolution network that extends existing image dense prediction architectures to 3D meshes. The novelty of our approach lies in two key aspects. First, we construct a multi-resolution mesh pyramid directly from the high-resolution input data and propose area-aware mesh downsampling/upsampling operations that use sequential bijective inter-surface mappings between different mesh resolutions. The inter-surface mapping redefines the mesh, rather than reshaping it, which thus avoids introducing unnecessary errors. Second, we maintain the high-resolution representation in the multi-resolution convolution network, enabling multi-scale fusions to exchange information across parallel multi-resolution subnetworks, rather than through connections of high-to-low resolution subnetworks in series. These features differentiate our approach from most existing mesh convolution networks and enable more accurate mesh dense predictions, which is confirmed in experiments.

Deformable registration of lung 3DCT images using an unsupervised heterogeneous multi-resolution neural network.

Lung image registration is more challenging than other organs. This is because the breath of the human body causes large deformations in the lung parenchyma and small deformations in tissues such as the pulmonary vascular. Many studies have recently used multi-resolution networks to solve the lung registration problem. However, they use the same structure of registration modules on each level, which makes it difficult to handle complex and small deformations. We propose an unsupervised heterogeneous multi-resolution network (UHMR-Net) to overcome the above problem. The image detail registration module (IDRM) is designed on the highest resolution level. Within this module, the cascaded network is used on the same resolution image to continuously learn the "remaining" detail deformation fields. The shallow shrinkage loss (SS-Loss) is designed to supervise the cascaded network, thus further improving the ability of the network to handle small deformations. Moreover, with the lightweight feature local correlation layer we proposed, the image boundary registration module (IBRM), on multiple low-resolution levels, can better solve the large deformation registration problem. The target registration error on the public DIR-Lab 4DCT dataset was 1.56 ± 1.39mm, which was significantly better than the classic conventional methods and advanced deep-based methods.

Lightweight multi-scale classification of chest radiographs via size-specific batch normalization

Background and Objective:Convolutional neural networks are widely used to detect radiological findings in chest radiographs. Standard architectures are optimized for images of relatively small size (for example, 224 × 224 pixels), which suffices for most application domains. However, in medical imaging, larger inputs are often necessary to analyze disease patterns. A single scan can display multiple types of radiological findings varying greatly in size, and most models do not explicitly account for this. For a given network, whose layers have fixed-size receptive fields, smaller input images result in coarser features, which better characterize larger objects in an image. In contrast, larger inputs result in finer grained features, beneficial for the analysis of smaller objects. By compromising to a single resolution, existing frameworks fail to acknowledge that the ideal input size will not necessarily be the same for classifying every pathology of a scan. The goal of our work is to address this shortcoming by proposing a lightweight framework for multi-scale classification of chest radiographs, where finer and coarser features are combined in a parameter-efficient fashion.Methods:We experiment on CheXpert, a large chest X-ray database. A lightweight multi-resolution (224 × 224, 448 × 448 and 896 × 896 pixels) network is developed based on a Densenet-121 model where batch normalization layers are replaced with the proposed size-specific batch normalization. Each input size undergoes batch normalization with dedicated scale and shift parameters, while the remaining parameters are shared across sizes. Additional external validation of the proposed approach is performed on the VinDr-CXR data set.Results:The proposed approach (AUC 83.27±0.17, 7.1M parameters) outperforms standard single-scale models (AUC 81.76±0.18, 82.62±0.11 and 82.39±0.13 for input sizes 224 × 224, 448 × 448 and 896 × 896, respectively, 6.9M parameters). It also achieves a performance similar to an ensemble of one individual model per scale (AUC 83.27±0.11, 20.9M parameters), while relying on significantly fewer parameters. The model leverages features of different granularities, resulting in a more accurate classification of all findings, regardless of their size, highlighting the advantages of this approach.Conclusions:Different chest X-ray findings are better classified at different scales. Our study shows that multi-scale features can be obtained with nearly no additional parameters, boosting performance.

Bipartite networks to enable management of internet core and metro infrastructure

In Telecommunications, core and metro network infrastructure provide high-speed, low-latency internet service to national-level areas. Their optimal operation and management are essential for today's increasing internet service demand. The usual representation of such infrastructure as a complex network considers router stations and other main internet sites as network nodes, connected by links that represent fibre-optic cables. However, this representation comes with important simplifications. One of them is that two nodes are connected by one single link when in reality there are involved multiple links in that connection. This paper explores bipartite networks to model the detail of multiple links connecting two network nodes, zooming-in process. To avoid dimensional complexities, the cable disentangles will only be in place at local sub-networks where the analysis may have a particular focus of interest. For instance, this process will support the decision-making associated to re-routing operations after a traffic anomaly has been detected. To show the efficiency of this multi-resolution network approach, the paper uses the case study of the core and metro infrastructure of one of the major UK internet service providers.

HMRNet: High and Multi-Resolution Network With Bidirectional Feature Calibration for Brain Structure Segmentation in Radiotherapy.

Accurate segmentation of Anatomical brain Barriers to Cancer spread (ABCs) plays an important role for automatic delineation of Clinical Target Volume (CTV) of brain tumors in radiotherapy. Despite that variants of U-Net are state-of-the-art segmentation models, they have limited performance when dealing with ABCs structures with various shapes and sizes, especially thin structures (e.g., the falx cerebri) that span only few slices. To deal with this problem, we propose a High and Multi-Resolution Network (HMRNet) that consists of a multi-scale feature learning branch and a high-resolution branch, which can maintain the high-resolution contextual information and extract more robust representations of anatomical structures with various scales. We further design a Bidirectional Feature Calibration (BFC) block to enable the two branches to generate spatial attention maps for mutual feature calibration. Considering the different sizes and positions of ABCs structures, our network was applied after a rough localization of each structure to obtain fine segmentation results. Experiments on the MICCAI 2020 ABCs challenge dataset showed that: 1) Our proposed two-stage segmentation strategy largely outperformed methods segmenting all the structures in just one stage; 2) The proposed HMRNet with two branches can maintain high-resolution representations and is effective to improve the performance on thin structures; 3) The proposed BFC block outperformed existing attention methods using monodirectional feature calibration. Our method won the second place of ABCs 2020 challenge and has a potential for more accurate and reasonable delineation of CTV of brain tumors.

An Efficient Multiresolution Network for Vehicle Reidentification

In general, vehicle images have varying resolutions due to vehicles’ movements and different camera settings. However, most existing vehicle re-identification models are single-resolution deep networks trained with pre-uniformly resizing vehicle images, which underestimates adverse effects of varying resolutions and leads to unsatisfactory performance. A straightforward solution for dealing with varying resolutions is to train multiple vehicle re-identification models. Each model is independently trained with images of a specific resolution. However, this straightforward solution requires significant overhead and ignores intrinsic associations among different resolution images. For that, an efficient multi-resolution network (EMRN) is proposed for vehicle re-identification in this paper. Firstly, EMRN embeds a newly-designed multi-resolution feature dimension uniform module (MR-FDUM) behind a traditional backbone network (i.e., ResNet-50). As a result, the whole model can extract fixed dimensional features from different resolution images so that it can be trained with one loss function of fixed dimensional parameters rather than training multiple models. Secondly, a multi-resolution image randomly feeding strategy is designed to train EMRN, making each mini-batch data of a random resolution during the training process. Consequently, EMRN can implicitly learn collaborative multi-resolution features via only a unitary deep network. Experiments on three large scale datasets, i.e., VeRi776, VehicleID, and VRIC, demonstrate that EMRN is superior to state-of-the-art vehicle re-identification methods.

A multiresolution network architecture for deferred neural lighting

AbstractWe present a novel multiresolution network architecture for deferred neural lighting. The key idea is to explicitly separate the processing of appearance at different spatial resolutions, leading to considerably improved high‐frequency details as well as temporal stability in animation sequences with varying view conditions. Moreover, our network is only half the size of the original one, and requires less training data to converge to satisfactory results. The network is tested over five captured datasets from deferred neural lighting and may be extended to other neural appearance techniques, such as NeRF or neural textures.

DRMNet: Difference Image Reconstruction Enhanced Multiresolution Network for Optical Change Detection

DRMNet: Difference Image Reconstruction Enhanced Multiresolution Network for Optical Change Detection

Factorized multi-scale multi-resolution residual network for single image deraining

Factorized multi-scale multi-resolution residual network for single image deraining

Functional magnetic resonance imaging progressive deformable registration based on a cascaded convolutional neural network.

Intersubject registration of functional magnetic resonance imaging (fMRI) is necessary for group analysis. Accurate image registration can significantly improve the results of statistical analysis. Traditional methods are achieved by using high-resolution structural images or manually extracting functional information. However, structural alignment does not necessarily lead to functional alignment, and manually extracting functional features is complicated and time-consuming. Recent studies have shown that deep learning-based methods can be used for deformable image registration. We proposed a deep learning framework with a three-cascaded multi-resolution network (MR-Net) to achieve deformable image registration. MR-Net separately extracts the features of moving and fixed images via a two-stream path, predicts a sub-deformation field, and is cascaded three times. The moving and fixed images' deformation field is composed of all sub-deformation fields predicted by the MR-Net. We imposed large smoothness constraints on all sub-deformation fields to ensure their smoothness. Our proposed architecture can complete the progressive registration process to ensure the topology of the deformation field. We implemented our method on the 1000 Functional Connectomes Project (FCP) and Eyes Open Eyes Closed fMRI datasets. Our method increased the peak t values in six brain functional networks to 19.8, 17.8, 15.0, 16.4, 17.0, and 13.2. Compared with traditional methods [i.e., FMRIB Software Library (FSL) and Statistical Parametric Mapping (SPM)] and deep learning networks [i.e., VoxelMorph (VM) and Volume Tweening Network (VTN)], our method improved 47.58%, 11.88%, 18.60%, and 15.16%, respectively. Our three-cascaded MR-Net can achieve statistically significant improvement in functional consistency across subjects.

Hierarchical and Multi-resolution Neuron Network Computing for Complex Inverse Problems-An Approach Based on Brain-inspired Computing and Learning Method

Background: Complex inverse problems such as Radar Imaging and CT/EIT imaging are well investigated in mathematical algorithms with various regularization methods. However it is difficult to obtain stable inverse solutions with fast convergence and high accuracy at the same time due to the ill-posed property and non-linear property. Objective: In this paper, we propose a hierarchical and multi-resolution scalable method from both algorithm perspective and hardware perspective to achieve fast and accurate solu-tions for inverse problems by taking radar and EIT imaging as examples. Method: We present an extension of discussion on neuromorphic computing as brain-inspired computing method and the learning/training algorithm to design a series of problem specific AI “brains” (with different memristive values) to solve a general complex ill-posed inverse problems that are traditionally solved by mathematical regular operators. We design a hierarchical and multi-resolution scalable method and an algorithm framework to train AI deep learning neuron network and map into the memristive circuit so that the memristive val-ues are optimally obtained. We propose FPGA as an emulation implementation for neuro-morphic circuit as well. Result: We compared the methodology between our approach and traditional regulariza-tion method. In particular we use Electrical Impedance Tomography (EIT) and Radar imaging as typical examples to compare how to design an AI deep learning neuron network architec-tures to solve inverse problems. Conclusion: With EIT imaging as a typical example, we show that any moderate complex inverse problem, as long as it can be described as combinational problem, AI deep learning neuron network is a practical alternative approach to try to solve the inverse problems with any given expected resolution accuracy, as long as the neuron network width is large enough and computational power is strong enough for all combination samples training purpose.