Granular Features Research Articles

Confidence-Induced Granular Partial Label Feature Selection via Dependency and Similarity

Confidence-Induced Granular Partial Label Feature Selection via Dependency and Similarity

Composition analysis of β-(In x Ga1-x )2O3 thin films coherently grown on (010) β-Ga2O3 via mist CVD

ABSTRACT This study investigates the compositional analysis and growth of β-(In x Ga1-x )2O3 thin films on (010) β-Ga2O3 substrates using mist chemical vapor deposition (CVD), including the effects of the growth temperature. We investigated the correlation between In composition and b-axis length in coherently grown films, vital for developing high-electron-mobility transistors and other devices based on β-(In x Ga1-x )2O3. Analytical techniques, including X-ray diffraction (XRD), reciprocal space mapping, and atomic force microscopy, were employed to evaluate crystal structure, strain relaxation, and surface morphology. The study identified a linear relationship between In composition and b-axis length in coherently grown films, facilitating accurate composition determination from XRD peak positions. The films demonstrated high surface flatness with root-mean-square roughness below 0.6 nm, though minor relaxation and granular features emerged at higher In compositions (x = 0.083) at the growth temperature of 750°C. XRD results revealed that lattice relaxation were observed at a growth temperature of 700°C despite low In composition. In contrast, at 800°C, the In composition was higher than at 750°C, and coherent growth was achieved. The surface morphology was the flattest at 750°C. These findings indicate that the growth temperature plays a crucial role in the mist CVD growth of β-(In x Ga1-x )2O3 thin films. This study offers insights into the relationship between In composition and lattice parameters in coherently grown β-(In x Ga1-x )2O3 films, as well as the effect of growth conditions, contributing to the advancement of ultra-wide bandgap semiconductor device development.

ScDTL: enhancing single-cell RNA-seq imputation through deep transfer learning with bulk cell information.

The increasing single-cell RNA sequencing (scRNA-seq) data enable researchers to explore cellular heterogeneity and gene expression profiles, offering a high-resolution view of the transcriptome at the single-cell level. However, the dropout events, which are often present in scRNA-seq data, remaining challenges for downstream analysis. Although a number of studies have been developed to recover single-cell expression profiles, their performance may be hindered due to not fully exploring the inherent relations between genes. To address the issue, we propose scDTL, a deep transfer learning based approach for scRNA-seq data imputation by harnessing the bulk RNA-sequencing information. We firstly employ a denoising autoencoder trained on bulk RNA-seq data as the initial imputation model, and then leverage a domain adaptation framework that transfers the knowledge learned by the bulk imputation model to scRNA-seq learning task. In addition, scDTL employs a parallel operation with a 1D U-Net denoising model to provide gene representations of varying granularity, capturing both coarse and fine features of the scRNA-seq data. Finally, we utilize a cross-channel attention mechanism to fuse the features learned from the transferred bulk imputation model and U-Net model. In the evaluation, we conduct extensive experiments to demonstrate that scDTL could outperform other state-of-the-art methods in the quantitative comparison and downstream analyses.

Granular correlation-based label-specific feature augmentation for multi-label classification

Multi-label classification is an extension of single-label classification with generations of multi-output for unseen instances. Label correlation is an essential component in constructing multi-label classifiers. How to optimize the representation of label correlation while preserving the semantics of label-specific remains an uncertain issue. Instead of estimating label correlation by a holistic feature representation, we present an augmented label correlation model by generating multi-granularity label-specific features. Firstly, we devise a mixture distance measure to characterize the closeness of an instance by weighing the Pearson correlation coefficient with cosine similarity. Secondly, we explore the local label-specific relative discrimination by leveraging from both the instance-level and class-level correlation distribution within k nearest neighborhood. Finally, we conduct an information fusion strategy to comprehensively integrate the positive and the negative tendencies at the neighborhood level. Instances with salient positive tendency and compact neighborhood structure receive larger values while receiving smaller values with salient negative tendency and sparse neighborhood structure. With the concatenation of original features and augmented features, we examine the classification performance of the proposed granule correlation-based feature augmentation (GOFA) on well-established second-order multi-label classification methods. Extensive comparisons on thirteen benchmarks demonstrate the statistical superiority of GOFA over state-of-the-art multi-label classifications.

The evolution of granular surface structure and functional properties in rice starch during grain filling

The developmental changes in the granular surface structure and functional properties of starch during the entire grain filling period of rice (around 40 days) were investigated. The specific surface area of rice starch significantly decreased firstly then stabilized during growth due to increasing granular size. The pore volume decreased from 5.40 cm3/g at 6th day after anthesis (DAA-6) to 3.02 cm3/g (DAA-46). More starch granule-associated proteins (SGAPs) accumulated on the surface and in channels. Swelling power decreased by 46 %, whereas the flow behavior index (n) decreased by 32 % in upward curve during starch development from DAA-6 to DAA-30. Tan δ first dropped then remained steady at DAA 22–34 and lightly rebounded at the final stage, indicating that starch in the middle stage tended to have greater viscoelastic gel behavior at all sweeps. Mature starch showed lower in vitro hydrolysis rate and exhibited stronger enzymatic resistance. The results showed that granular surface features of rice starch may be an essential factor in determining rheological behavior and resistance to hydrolysis.

SAFLFusionGait: Gait recognition network with separate attention and different granularity feature learnability fusion

Gait recognition, an essential branch of biometric identification, uses walking patterns to identify individuals. Despite its effectiveness, gait recognition faces challenges such as vulnerability to changes in appearance due to factors like angles and clothing conditions. Recent progress in deep learning has greatly enhanced gait recognition, especially through methods like deep convolutional neural networks, which demonstrate impressive performance. However, current approaches often overlook the connection between coarse-grained and fine-grained features, thereby restricting their overall effectiveness. To address this limitation, we propose a new framework for gait recognition framework that combines deep-supervised fine-grained separation with coarse-grained feature learnability. Our framework includes the LFF module, which consists of the SSeg module for fine-grained information extraction and a mechanism for fusing coarse-grained features. Furthermore, we introduce the F-LCM module to extract local disparity features more effectively with learnable weights. Evaluation on CASIA-B and OU-MVLP datasets shows superior performance compared to classical networks.

EEG emotion recognition using EEG-SWTNS neural network through EEG spectral image

The rapid development in deep learning models provide considerable advancements in emotion recognition by electroencephalogram (EEG). However, existing approaches primarily focus on temporal and frequency domain features of EEG signals, neglecting spatial domain features in model information integration. Therefore, this study proposed a novel EEG input format, called EEG spectral image (ESI), which integrates spatial domain features using Azimuthal Equidistant Projection (AEP) and frequency domain features through differential entropy (DE). To better utilize this fine-grained data format, we propose the EEG Swin Transformer (EEG-SWTNS), which combines the window attention mechanism with shifted window partitioning. Window attention mechanism can concentrate on affective information within various regions and use shifted window partitioning to disrupt aggregated emotional representations for more granular features. Different from traditional graph neural networks, this method leverages the spatial locality of region information, leading to enhanced performance in EEG-based emotion recognition. Experiments conducted on SEED and SEED IV datasets demonstrate superior performance compared to baseline methods and the state-of-the-art models. Relative improvements of 0.6% and 0.08% are observed in subject-dependent experiments, while accuracies of 80.07% and 66.72% are achieved in subject-independent experiments without utilizing transfer learning techniques.

A shunted-swin transformer for surface defect detection in roller bearings

Although surface defect detection for roller bearings has received much attention in smart industrial manufacturing, most existing approaches employed the existing Convolution or Transformer models without modification to detect the defects. Such applications are undesirable because the classification models lack the necessary spatial relationship between multiple objects within an image. To develop a specialized model to detect bearing defects, we expand the applicability of the Swin Transformer. This paper presented a shunted-window Transformer (sSwin), which replaces the shifted-window mechanism with the shunted large–small windows. In sSwin, the self-attention heads can capture multi-scale token granularity features within the same self-attention layer. Also, a proposed local connection module enhances the boundary interaction of adjacent value tokens. As a result of the different-scale receptive fields of tokens within one attention layer, the model can bring significant performance gain on the defective detection tasks. The experiment demonstrates that sSwin-T outperforms the classical Swin-T by +2.8% APb and +1.9% APm, and Swinv2-T by +1.0% APb and +1.1% APm on COCO. It also surpasses them by +1.1% APb and +3.4% APb on the private bearing dataset with Cascade Mask-RCNN 3x.

Multi-granularity relationship reasoning network for high-fidelity 3D shape reconstruction

Monocular image-based 3D reconstruction is widely used in virtual reality, augmented reality, and autonomous driving, which benefits from the rapid development of deep learning approaches. Most of the available methods focused on reconstructing the overall shape of the object while ignoring some fine-grained details. Moreover, these methods make it hard to exactly reconstruct complex topological structures. In this paper, we propose a multi-granularity relationship reasoning network (MGRRNet), which aims to recover 3D shapes with high fidelity and rich details via the relationship reasoning between different granularity information. Specifically, our model captures the discriminative and detailed features at different granularities for extracting attentional regions. Then we perform the relationship reasoning between different granularities to reinforce the multi-granularity consistency and inter-granularity correlation. By doing this, our network is able to achieve robust feature representation and fine reconstruction. During the learning process, we jointly optimize procedures of different granularity feature representations via a sequence of inter-granularity cycle loss iterations. Extensive experimental results on two publicly available datasets justify that our approach achieves competitive performance compared to the state-of-the-art methods. Codes and all resources will be publicly available at https://github.com/Ray-tju/MGRRNet.

Text to image synthesis with multi-granularity feature aware enhancement Generative Adversarial Networks

Synthesizing complex images from text presents challenging. Compared to autoregressive and diffusion model-based methods, Generative Adversarial Network-based methods have significant advantages in terms of computational cost and generation efficiency yet remain two limitations: first, these methods often refine all features output from the previous stage indiscriminately, without considering these features are initialized gradually during the generation process; second, the sparse semantic constraints provided by the text description are typically ineffective for refining fine-grained features. These issues complicate the balance between generation quality, computational cost and inference speed. To address these issues, we propose a Multi-granularity Feature Aware Enhancement GAN (MFAE-GAN), which allows the refinement process to match the order of different granularity features being initialized. Specifically, MFAE-GAN (1) samples category-related coarse-grained features and instance-level detail-related fine-grained features at different generation stages based on different attention mechanisms in Coarse-grained Feature Enhancement (CFE) and Fine-grained Feature Enhancement (FFE) to guide the generation process spatially, (2) provides denser semantic constraints than textual semantic information through Multi-granularity Features Adaptive Batch Normalization (MFA-BN) in the process of refining fine-grained features, and (3) adopts a Global Semantics Preservation (GSP) to avoid the loss of global semantics when sampling features continuously. Extensive experimental results demonstrate that our MFAE-GAN is competitive in terms of both image generation quality and efficiency.

Range control-based class imbalance and optimized granular elastic net regression feature selection for credit risk assessment

Range control-based class imbalance and optimized granular elastic net regression feature selection for credit risk assessment

Status of dual-readout calorimetry for future high-energy physics experiments

Future experiments at high energy e + e - colliders will focus on extremely precise Standard Model measurements. Among the most important physics benchmarks, there is the capability to resolve the Higgs decays into W or Z pairs, in their completely hadronic decay modes (4 jets in the final state), only based on the invariant mass of the jet pair coming from decay of the on-shell boson. This translates into a relative energy resolution target of 30%/√E, well beyond current detector performances. Dual-readout calorimetry is a technique which aims to improve the energy resolution, for single hadrons and hadronic jets, exploiting the information produced by two different physical processes, namely scintillation and Čerenkov light emission. The IDEA detector, whose concept has been included in both the FCC and CEPC Conceptual Design Reports, is based on a dual-readout fibre calorimeter with independent fibre readout exploiting Silicon PhotoMultipliers (SiPMs). The individual SiPM information will be beneficial for a highly granular calorimeter design, opening up to advanced reconstruction techniques such as Particle Flow and a variety of neural network algorithms.In this paper the status of calorimeter prototypes that have been developed to demonstrate the feasibility of the dual-readout method in association with the high granularity feature is illustrated. The specific choice for the design of each prototype is presented, together with the performances achieved at high-energy test beams or through simulations.

The regional pervasiveness of local productivity shocks on macroeconomic output in Europe

ABSTRACT This paper studies inter-regional technology spillover in EU-28 countries and shows how regional total factor productivity (TFP) shocks shape aggregate productivity in Europe. This paper provides evidence that several regions play a significant role as common factors in the production function of almost every other region. We provide a procedure to identify and test these regions through inter-regional trade flows. In our estimations, the ten most pervasive regions out of 272 NUTS 2 regions explain about 12% of the total factor productivity growth in the EU-28 countries. Significant heterogeneity exists in the region’s ability to absorb imported knowledge among these countries. Local knowledge stock’s contribution is limited compared to imported knowledge due to a regional disparity in R&D distribution. Our work contributes to understanding the granular features of spatial macroeconomics, with a particular focus on inter-regional trade flows and knowledge spillovers.

Fabrication of Highly Sensitive YCeO Chemo-resistive Gas Sensor for Selective Detection of CO2

A room-temperature-operated CO2 gas sensor based on YCeO nanocomposite was effectively prepared by the simple hydrothermal technique to detect low traces of CO2 (50–250 ppm). The YCeO granular morphological features were observed using field-emission scanning electron microscopy, which confirmed successful fabrication of nanocomposite of Y2O3 and CeO2. X-ray diffraction of YCeO showed the Cubic structure of space group Fm3m having density 6.74 gmcm−3. Rietveld refinement was performed for the analysis of complete crystal structural property. Surface porosity and specific surface area were observed by Brunnauer-Emmet Teller analysis. Optical properties were observed using UV-Visible spectroscopy. The band gap, optical conductivity, and refractive index calculated were 3.44 eV, 2.63 × 106, and 0.1164, respectively. Fourier transform infrared spectroscopy was done to analyze the functional and elastic properties of as-prepared nanomaterial. The highest sensor response recorded was 2.14. The response and recovery time at 50 ppm observed were 75.6 and 107.3 s, respectively. The YCeO chemo-resistive sensor confirmed long-term stability and selectivity to CO2 as compared to other gases viz. LPG, NH3, CH4, H2S, NO2 and H2. The relative humidity exposure was also performed at 15, 55 and 95% RH, in which it was confirmed that the sensor would give best response at mid humidity level i.e. 55 %RH. Sensing characteristics curve of YCeO nanocomposite at different temperature (30 °C–90 °C) at 50 ppm confirmed that YCeO sensor performed excellent at room temperature. This report unlocks an innovative opening for the fabrication of sensing devices that are room-temperature-operatable, highly sensitive and selective for quick detection of CO2 gas for its commercialization.

Multi-branch residual image semantic segmentation combined with inverse weight gated-control

The loss of pixel-level information in the multi-class segmentation task based on the U-net model results in unclear boundaries and low semantic segmentation accuracy. Aiming at this, a deep multi-branch residual Unet (IWG-MRUN) with fused inverse weight gated-control is proposed to improve the quality of image semantic segmentation. Specifically, we first introduce a deep multi-branch residual module, which used parallel convolution mode to capture the contextual feature to extract the detailed features of the input image at a deeper level. Then, we adopt an inverse weight gated-control module to enhance the diversity of up-sampling information by counterclockwise transmitting attention horizontally to improve the restoration accuracy of up-sampled image pixels. Finally, to obtain finer granularity features from low spatial resolution images, we adopt the different receptive field pyramid attention mechanisms at the highest level of the U-shaped encoder to capture high-level context information at different scales, thereby improving the accuracy of semantic segmentation. The experimental results show that the segmentation accuracy of the proposed algorithm reaches 91.80% and the CCE loss is reduced to 0.21. When compared to the Unet, BiSeNet, DeeplabV3+ and U-net + BLR model, the pixel accuracy of semantic segmentation is improved by 15.0%, 1.98%, 0.9% and 6.5%, respectively. The semantic segmentation model proposed in this paper provides an end-to-end semantic segmentation capability with the enriched finer granularity features of the target boundary and realizes the accurate segmentation of the objects in different categories.

Cross-modal Multiple Granularity Interactive Fusion Network for Long Document Classification

Long Document Classification (LDC) has attracted great attention in Natural Language Processing and achieved considerable progress owing to the large-scale pre-trained language models. In spite of this, as a different problem from the traditional text classification, LDC is far from being settled. Long documents, such as news and articles, generally have more than thousands of words with complex structures. Moreover, compared with flat text, long documents usually contain multi-modal content of images, which provide rich information but not yet being utilized for classification. In this article, we propose a novel cross-modal method for long document classification, in which multiple granularity feature shifting networks are proposed to integrate the multi-scale text and visual features of long documents adaptively. Additionally, a multi-modal collaborative pooling block is proposed to eliminate redundant fine-grained text features and simultaneously reduce the computational complexity. To verify the effectiveness of the proposed model, we conduct experiments on the Food101 dataset and two constructed multi-modal long document datasets. The experimental results show that the proposed cross-modal method outperforms the single-modal text methods and defeats the state-of-the-art related multi-modal baselines.

A novel dual-granularity lightweight transformer for vision tasks

Transformer-based networks have revolutionized visual tasks with their continuous innovation, leading to significant progress. However, the widespread adoption of Vision Transformers (ViT) is limited due to their high computational and parameter requirements, making them less feasible for resource-constrained mobile and edge computing devices. Moreover, existing lightweight ViTs exhibit limitations in capturing different granular features, extracting local features efficiently, and incorporating the inductive bias inherent in convolutional neural networks. These limitations somewhat impact the overall performance. To address these limitations, we propose an efficient ViT called Dual-Granularity Former (DGFormer). DGFormer mitigates these limitations by introducing two innovative modules: Dual-Granularity Attention (DG Attention) and Efficient Feed-Forward Network (Efficient FFN). In our experiments, on the image recognition task of ImageNet, DGFormer surpasses lightweight models such as PVTv2-B0 and Swin Transformer by 2.3% in terms of Top1 accuracy. On the object detection task of COCO, under RetinaNet detection framework, DGFormer outperforms PVTv2-B0 and Swin Transformer with increase of 0.5% and 2.4% in average precision (AP), respectively. Similarly, under Mask R-CNN detection framework, DGFormer exhibits improvement of 0.4% and 1.8% in AP compared to PVTv2-B0 and Swin Transformer, respectively. On the semantic segmentation task on the ADE20K, DGFormer achieves a substantial improvement of 2.0% and 2.5% in mean Intersection over Union (mIoU) over PVTv2-B0 and Swin Transformer, respectively. The code is open-source and available at: https://github.com/ISCLab-Bistu/DGFormer.git.

Partial multi-label learning via robust feature selection and relevance fusion optimization

Partial Multi-Label Learning (PML) is a more practical learning paradigm, in which the labeling information is ambiguated. Most existing PML algorithms rely on assumptions to resolve ambiguity. However, these assumptions do not account for the origin of the noise labeling and therefore fail to address the impact of noise on the learner’s performance at the root. In this paper, we will propose a PML method jointly granular ball-based robust feature selection and relevance fusion optimization (PML-GR). Specifically, in the first stage, we construct a granular ball to compute the core-set with weights and then design a feature importance evaluation function to assign weights to each feature in the core-set, resulting in a ranking of feature importance for the PML learner; in the second stage, based on the selected features, a fusion-based objective function is constructed to compute the label confidence by taking into account the joint effect of the global sample similarity and local label relevance. Finally, a multi-label prediction model is learned by fitting the multi-output regressor to the label confidence. The experimental results demonstrate that the proposed method achieves competitive generalization performance by effective feature selection and relevance fusion optimization, which can focus more on discriminative features and minimize the effect of noisy labels during training.

Deep CNN Based Multi Object Detection And Tracking In Video Frames With Mean Distributed Feature Set

The computer vision field of object identification and tracking has been put to use in a broad variety of settings, from medical monitoring to autonomous vehicle navigation to anomaly detection and beyond. The efficiency of object detectors and trackers has substantially increased with the rapid growth of Deep Learning (DL) networks and the computational power of GPUs. Due to their poor generalization performance, classic extracting feature models are limited in their ability to recognize many targets in complicated scenarios, relying instead on low-level feature information such as contour information and texture information. In addition to gleaning granular texture features from pre-level Convolution Neural Networks (CNN), these models trained with deep learning can also glean meta-level knowledge from the post-level convolution layer. This research proposes a Deep CNN based Multi Object Detection and Tracking in Video Frame with Mean Distributed Feature Set (DCNN-MODT-MDFS) model for automatically assigning input objects to one of several predefined categories. Following object detection and tracking, the suggested techniques assign a class label by analyzing numerous frames in a video. The images could be taken from the video frames that are used to train the model. In order to account for the disparity in aspect ratio, the proposed method employs multi-scale feature maps for object detection and uses filters with unique default boxes. The model is trained until there is a noticeable drop in the error rate. The proposed model multi object detection rate and tracking rate is contrasted with the existing models and the results represent that proposed model performance is high.

A light-weight model with granularity feature representation for fine-grained visual classification

A light-weight model with granularity feature representation for fine-grained visual classification