Semantic Network Research Articles

Structure-Guided Image Inpainting Based on Multi-Scale Attention Pyramid Network

Current single-view image inpainting methods often suffer from low image information utilization and suboptimal repair outcomes. To address these challenges, this paper introduces a novel image inpainting framework that leverages a structure-guided multi-scale attention pyramid network. This network consists of a structural repair network and a multi-scale attention pyramid semantic repair network. The structural repair component utilizes a dual-branch U-Net network for robust structure prediction under strong constraints. The predicted structural view then serves as auxiliary information for the semantic repair network. This latter network exploits the pyramid structure to extract multi-scale features of the image, which are further refined through an attention feature fusion module. Additionally, a separable gated convolution strategy is employed during feature extraction to minimize the impact of invalid information from missing areas, thereby enhancing the restoration quality. Experiments conducted on standard datasets such as Paris Street View and CelebA demonstrate the superiority of our approach over existing methods through quantitative and qualitative comparisons. Further ablation studies, by incrementally integrating proposed mechanisms into a baseline model, substantiate the effectiveness of our multi-view restoration strategy, separable gated convolution, and multi-scale attention feature fusion.

키워드 네트워크 분석을 활용한 인공지능 융합 영재교육 연구 동향 분석

Objectives The purpose of this study is to analyze trends in artificial intelligence convergence gifted education in elementary and middle schools that have been studied with a focus on artificial intelligence, and to derive important issues, topics, trends, and trends in related research. Methods Research trends were analyzed for 75 literatures related to artificial intelligence and artificial intelligence convergence gifted education research and education programs. Analysis data was collected through Academic Research Information Service (RISS). The collected data was refined using Excel and text, and major keywords were identified, analyzed, and visualized through frequency analysis and keyword network analysis. CONCOR cluster analysis was used to determine the centrality of keywords. Results First, since 2017, research on artificial intelligence convergence gifted education has been steadily expanding in terms of content and methods. Second, the results of frequency analysis of the main keywords in research related to artificial intelligence convergence gifted education showed ‘artificial intelligence’, ‘artificial intelligence education’, ‘gifted education’, ‘computational thinking ability’, and ‘machine learning’.Third, as a result of TF-IDF weight analysis, ‘artificial intelligence’, ‘artificial intelligence education’, ‘gifted education’, ‘computational thinking ability’, ‘science gifted’, ‘artificial intelligence education program’, and ‘information gifted’ were found. In general, it was similar to the results of the frequency analysis of major keywords, but it was confirmed that there was a difference in the rankings. Keywords with relatively high weight compared to their frequency of occurrence are ‘science gifted’, ‘creativity’, and ‘artificial intelligence education program’.Fourth, through semantic network and CONCOR analysis, it can be confirmed that artificial intelligence convergence gifted education research is being conducted focusing on ‘educational content’, ‘educational effect’, ‘teaching method’, and ‘educational target’. Conclusions This study derived the main topics and flows of research based on the results of keyword network analysis of domestic artificial intelligence convergence gifted education-related research.

Unmasking colorectal cancer: A high-performance semantic network for polyp and surgical instrument segmentation

Colorectal cancer (CRC) remains a significant health concern, with colonoscopy serving as the gold standard for diagnosis. Accurately segmenting polyps from colonoscopy images is crucial for detecting polyps and preventing CRC. However, challenges such as varying polyp sizes, blurred edges, and uneven brightness hinder segmentation accuracy. Leveraging artificial intelligence (AI) and robot-assisted surgery mechanisms can aid surgeons and physicians in detecting and treating polyps. To address these challenges, we propose a Colorectal Network (CR-Net), an AI-based encoder-decoder network for precise polyp and surgical instrument segmentation. CR-Net incorporates a pre-trained Visual Geometry Group model with 16 convolution layers (VGG16), attention mechanisms, redesigned skip connections, and horizontal dense connections within a U-Net architecture. The VGG16 encoder captures robust visual features, while redesigned skip connections accommodate complex data dimensions, leading to enhanced segmentation outcomes. Horizontal dense connections transfer overlooked features from the encoder to subsequent layers, further improving segmentation accuracy. Additionally, a spatial attention block enhances spatial features and ensures compatibility during upsampling. Evaluation of datasets including the Kvasir segmentation (Kvasir-SEG) dataset, Computer Vision Center Clinic Database (CVC-ClinicDB), Kvasir-Instrument dataset, and University of Washington Sinus Surgery Live (UW-Sinus-Surgery-Live) dataset demonstrates CR-Net's superior performance, achieving Dice Similarity Coefficients of 96.21%, 96.54%, 96.32%, and 92.84%, respectively, surpassing previous methods. These results highlight CR-Net's potential in empowering healthcare professionals through advanced AI-driven engineering applications. By bridging AI techniques with engineering innovations, CR-Net represents a significant advancement in CRC diagnosis and treatment.

Weakly supervised semantic segmentation by knowledge graph inference

The weakly supervised semantic segmentation (WSSS) training based on image-level labels in convolutional neural network (CNN) is usually divided into two stages: multi-label classification and semantic segmentation. However, most of the existing work focuses on the improvement of the multi-label classification network stage, and little effort has been done to improve the performance of the downstream segmentation networks. In addition, CNN-based local convolution lacks in modeling extensive dependencies among categories. Therefore, in this paper, we propose a graph reasoning method to improve both the upstream and the downstream stages of the multi-label classification network and the semantic segmentation networks. In the multi-label classification network, we utilize external knowledge combined with a graph convolutional network (GCN) to perform global reasoning on the dependencies of each category. In the segmentation network, the Graph Reasoning Mapping Module (GRM) is proposed to explore the knowledge acquired from text corpora and facilitate contextual reasoning in various categories of image regions. The proposed GRM module is able to enhance the feature representation of local convolutions on the high-level semantics of the segmentation network, and adaptively learn the semantic consistency of each sample. We achieve state-of-the-art WSSS performance on PASCAL VOC 2012 and MS-COCO 2014 datasets with only image-level supervision. Extensive experiments on multi-label classification networks and semantic segmentation networks demonstrate the effectiveness of our proposed graph reasoning method on WSSS. Our code is available at:https://github.com/JIA-ZHANG666/GRM_layer.

Enhanced damage segmentation in RC components using pyramid Haar wavelet downsampling and attention U-net

Damage identification in post-earthquake reinforced concrete (RC) structures based on semantic segmentation has been recognized as a promising approach for rapid and non-contact damage localization and quantification. In damage segmentation tasks, damage regions are often set against complex backgrounds, featuring irregular geometric boundaries and intricate textures, posing significant challenges to model segmentation performance. Additionally, the absence of public datasets exacerbates these challenges, hindering advancements in this field. In this paper, a pyramid Haar wavelet downsampling attention UNet (PHA-UNet) semantic segmentation network is proposed, and a database containing 1400 images of damaged RC components (PEDRC-Dataset) with pixel-level annotations is established. In the proposed PHA-UNet, attention mechanisms, multiscale feature fusion, Haar wavelet downsampling, and transfer learning are introduced to address above challenges. Finally, the proposed PHA-UNet is compared with four existing image segmentation architectures on both the Cityspace and the PEDRC-Dataset.

Sharing is caring? How moral foundation frames drive the sharing of corrective messages and misinformation about COVID-19 vaccines

Drawing from Moral Foundation Theory, our study explores if and how corrective messages and misinformation related to COVID-19 vaccines utilize moral frames. Unlike studies that either focused on content-analyzing messages or study how audiences react to moral frames, this study incorporated both a content analysis of COVID vaccine messages and modeling of how millions of audiences reacted to such messages. We combined semantic network analysis, text-mining, and machine learning to analyze a large corpus of Facebook posts about COVID-19 vaccines. Our results showed that both corrective messages and misinformation prevalently deployed moral framing. We also found that while corrective messages tend to highlight the virtuous aspect of morality, misinformation focuses on the sinful aspect. In both contexts, the five moral frames could construct logically self-consistent worldviews. Moreover, for corrective messages, fairness, sanctity, care, authority, and loyalty frames all significantly influence users’ message sharing. For misinformation, only the authority/subversion frame was influential.

Fatigue Crack Detection Based on Semantic Segmentation Using DeepLabV3+ for Steel Girder Bridges

Artificial intelligence technology is receiving more and more attention in structural health monitoring. Fatigue crack detection in steel box girders in long-span bridges is an important and challenging task. This paper presents a semantic segmentation network model for this task based on DeepLabv3+, ResNet50, and active learning. Specifically, the classification network ResNet50 is re-tuned using the crack image dataset. Secondly, with the re-tuned ResNet50 as the backbone network, a crack semantic segmentation network was constructed based on DeepLabv3+, which was trained with the assistance of active learning. Finally, optimization for the probability threshold of the pixel category was performed to improve the pixel-level detection accuracy. Tests show that, compared with the crack detection network based on conventional ResNet50, this model can improve MIoU from 0.6181 to 0.7241.

RFSC-net: Re-parameterization forward semantic compensation network in low-light environments

Although detectors currently perform well in well-light conditions, their accuracy decreases due to insufficient object information. In addressing this issue, we propose the Re-parameterization Forward Semantic Compensation Network (RFSC-Net). We propose the Reparameterization Residual Efficient Layer Aggregation Networks (RSELAN) for feature extraction, which integrates the concepts of re-parameterization and the Efficient Layer Aggregation Networks (ELAN). While focusing on the fusion of feature maps of the same dimension, it also incorporates upward fusion of lower-level feature maps, enhancing the detailed texture information in higher-level features. Our proposed Forward Semantic Compensation Feature Fusion (FSCFF) network reduces interference from high-level to low-level semantic information, retaining finer details to improve detection accuracy in low-light conditions. Experiments on the low-light ExDark and DarkFace datasets show that RFSC-Net improves mAP by 2% on ExDark and 0.5% on DarkFace over the YOLOv8n baseline, without an increase in parameter counts. Additionally, AP50 is enhanced by 2.1% on ExDark and 1.1% on DarkFace, with a mere 3.7 ms detection latency on ExDark.

CloudSwinNet: A hybrid CNN-transformer framework for ground-based cloud images fine-grained segmentation

Solar irradiance is the main factor affecting the output of a photovoltaic (PV) power station. The dominant role of ground-based clouds on the variation of direct solar radiation in the whole sky. Ground-based cloud segmentation plays a crucial role in photovoltaic power generation prediction. Despite advancements, current cloud segmentation methods fail to meet the requirements of this application. While convolutional neural networks demonstrate impressive segmentation capabilities in ground-based cloud segmentation tasks, their inherent limitations hinder further performance enhancement. Hence, this paper introduces CloudSwinNet, a hybrid CNN-Transformer framework for ground-based cloud images fine-grained segmentation. CloudSwinNet leverages concepts from convolutional neural networks, including stepwise downsampling, local convolution, and skip connections. To further explore the fine-grained features in the ground-based cloud images, we incorporates the Fine-grained Feature Fusion Module (Fg-FFM) in encoder structure, while the jump connection structure integrates the GloRe spatial graph inference module. These additions enable comprehensive learning of multi-scale and long-range dependencies. We conducted extensive experiments on the fine-grained ground-based cloud dataset, demonstrating that CloudSwinNet outperforms six semantic segmentation networks in both qualitative and quantitative evaluations. The results of ablation experiments prove the effectiveness of the module introduced in this paper.

SFA-Net: Semantic Feature Adjustment Network for Remote Sensing Image Segmentation

Advances in deep learning and computer vision techniques have made impacts in the field of remote sensing, enabling efficient data analysis for applications such as land cover classification and change detection. Convolutional neural networks (CNNs) and transformer architectures have been utilized in visual perception algorithms due to their effectiveness in analyzing local features and global context. In this paper, we propose a hybrid transformer architecture that consists of a CNN-based encoder and transformer-based decoder. We propose a feature adjustment module that refines the multiscale feature maps extracted from an EfficientNet backbone network. The adjusted feature maps are integrated into the transformer-based decoder to perform the semantic segmentation of the remote sensing images. This paper refers to the proposed encoder–decoder architecture as a semantic feature adjustment network (SFA-Net). To demonstrate the effectiveness of the SFA-Net, experiments were thoroughly conducted with four public benchmark datasets, including the UAVid, ISPRS Potsdam, ISPRS Vaihingen, and LoveDA datasets. The proposed model achieved state-of-the-art accuracy on the UAVid, ISPRS Vaihingen, and LoveDA datasets for the segmentation of the remote sensing images. On the ISPRS Potsdam dataset, our method achieved comparable accuracy to the latest model while reducing the number of trainable parameters from 113.8 M to 10.7 M.

Application of Generative Adversarial Networks in Semi-Annotated Defect Synthesis and Detection Under Limited Resources

Defect detection is a crucial technology that is extensively employed in the manufacturing industry to monitor and ensure the quality of output. Deep learning models have shown remarkable potential for defect detection. However, the success of these models heavily relies on voluminous training data. Collecting substantial amounts of defect data is challenging in practical settings, and the tedious process of pixel-level defect annotation further complicates the task. Among the common defects encountered in manufacturing, scratches are particularly significant. To address these challenges, this study proposes a two-phase generative adversarial network (GAN) approach for synthesizing defect images and generating semi-automatic pixel-wise labels for anomaly detection. The first phase primarily focuses on synthesizing images, while the second phase involves the pixel-wise labeling of the images. The synthesized paired images generated by the GANs serve as input to the semantic network. Notably, the proposed methodology requires only a few real defect samples for training and a small amount of annotated data, making it practical and computationally efficient for implementation in the manufacturing industry. Experimental results indicate the effectiveness of the proposed deep-learning solution in defect detection, specifically in identifying scratches on various textured and patterned surfaces. A notably high detection accuracy is achieved, validating the potential of the approach in real-world manufacturing scenarios.

The role of animacy in language production: evidence from bare noun naming.

According to Levelt's language production model, to name an object, speakers must first conceptualise and lexicalise the object before its name can be articulated. Conceptualisation is conducted through the semantic network that exists at the conceptual level, with the highly activated concept(s) activating lexical items at the lemma level, that is, lexicalisation. So far, research focused mostly on semantic categories (i.e., semantic interference) but less so on animacy-a concept that is correlated with semantic categories. To investigate the role of this semantic feature in language production, we conducted a picture-word interference study in Mandarin Chinese, varying animacy congruency and controlling for classifier congruency while recording behavioural and electrophysiological responses. We observed an animacy interference effect together with a larger N400 component for animacy-incongruent versus congruent picture-word pairs, suggesting animacy-congruent concepts may be in closer proximity and hence lead to a stronger spreading of activation relative to animacy-incongruent concepts. Furthermore, a larger P600 component was observed for classifier-incongruent versus classifier-congruent picture-word pairs, suggesting syntactically driven processing of classifiers at the lemma level.

A semi-supervised framework fusing multiple information for knowledge graph entity alignment

Entity alignment (EA) is a fundamental task for cross linguistic knowledge graphs (KGs) understanding and interaction, which is committed to matching entities from different graphs based on their inherent semantics. Methods based on graph neural networks (GNNs) dominate the EA task, however, the majority of them ignore the higher-order information among entities in the KGs. Meanwhile, as important auxiliary information, the relational semantics, string information of entity names and attribute information of entities are insufficiently exploited during the inference phase. In addition, labeled alignment data is universally insufficient across various datasets, which limits the performance of the model. In this paper, we propose a Semi-supervised EA framework that Comprehensively considers both Structural and Attribute information within KGs (SCSA) to address these problems above. Specifically, our approach first leverages hypergraph neural networks (HGNN) to aggregate relational semantic information and graph convolutional networks (GCNs) with a highway filtering strategy to acquire the embedding representation of entities precisely. Then, we propose a bidirectional filtering technique with a combination of entity, attribute and string values to create pseudo-labeled data and lead the model for iteratively training. We implement our proposed framework on several publicly recognized cross-lingual datasets. The experimental results indicate that our framework outperforms almost all state-of-the-art (SOTA) methods.

The relationship between creativity and language as measured by linguistic maturity and text production

Creativity is the result of several converging elements, including cultural, environmental, and social factors. The context in which people live their lives not only affects their creativity; it also affects how they acquire and use language. This study shows how a person's context and linguistic development, as measured by linguistic maturity, affects their creativity. 249 students aged between 13 and 15, participated in the study by completing an online test measuring their linguistic maturity and creativity. This study used Guilford's (1967) Alternative Uses Test following the Consensual Assessment Technique. Regression models revealed a positive and significant influences of linguistic maturity on creativity. Greater semantic memory networks allow for more efficient searches, thus enabling creativity. Therefore, greater semantic richness leads to greater linguistic maturity, resulting in increased creativity. Our findings confirm a relationship between language skills and creativity (i.e., linguistic maturity and text production) at a deeper level than previously studied. This is relevant for both curriculum development and instructional design. In this sense, tools as ChatGPT that produce text from millions of documents, push the student to be mediator of all this linguistic richness. Furthermore, measuring linguistic maturity and text production provides another means to measure creativity, transcendent, considering that having proper approaches to measure creativity is a challenge.

The social valuation of a free-flowing river: A comparative analysis between two communities using semantic networks

Los ríos de flujo libre (RFL) se caracterizan por su conectividad fluvial íntegra y por proveer múltiples beneficios a las comunidades asentadas en ellos. Sin embargo, a nivel mundial estos tipos de ríos se encuentran amenazados por la construcción de obras de infraestructura. En ocasiones, la pluralidad de valores que las comunidades le atribuyen a estos ríos no se toma en cuenta, particularmente al momento de planificar proyectos de desarrollo, lo que podría detonar conflictos ambientales. Se analizaron los valores sociales asociados con el río Los Horcones (RLH) entre dos localidades, río arriba (Las Juntas y Los Veranos) y río abajo (Boca de Tomatlán). A partir del análisis de 35 entrevistas y el uso redes semánticas, se encontró que ambas comunidades tienen un interés común por la protección y conservación del río Los Horcones. Sin embargo, se observa una ponderación diferencial de valores específicos por localidad. Río abajo (Boca de Tomatlán) predominan los valores intrínsecos y relacionales, dado que el RLH es fundamental para la diversificación de sus medios de vida. En contraste, río arriba (Las Juntas y Los Veranos) predominan los valores relacionales e instrumentales debido a la dependencia económica del turismo de naturaleza que se realiza en el RLH. El reconocimiento de la pluralidad de valores asociados con los ríos de flujo libre es esencial para la formulación de estrategias de manejo y conservación de los recursos hídricos.

Neural 3D Scene Reconstruction With Indoor Planar Priors.

This paper addresses the challenge of reconstructing 3D indoor scenes from multi-view images. Many previous works have shown impressive reconstruction results on textured objects, but they still have difficulty in handling low-textured planar regions, which are common in indoor scenes. An approach to solving this issue is to incorporate planar constraints into the depth map estimation in multi-view stereo-based methods, but the per-view plane estimation and depth optimization lack both efficiency and multi-view consistency. In this work, we show that the planar constraints can be conveniently integrated into the recent implicit neural representation-based reconstruction methods. Specifically, we use an MLP network to represent the signed distance function as the scene geometry. Based on the Manhattan-world assumption and the Atlanta-world assumption, planar constraints are employed to regularize the geometry in floor and wall regions predicted by a 2D semantic segmentation network. To resolve the inaccurate segmentation, we encode the semantics of 3D points with another MLP and design a novel loss that jointly optimizes the scene geometry and semantics in 3D space. Experiments on ScanNet and 7-Scenes datasets show that the proposed method outperforms previous methods by a large margin on 3D reconstruction quality.

Intelligent system based on multiple networks for accurate ovarian tumor semantic segmentation

Ovarian tumors, especially malignant ones, represent a global concern, with increased prevalence in recent years. More accurate medical support systems are urgently needed to support medical staff in obtaining an efficient ovarian tumors diagnosis since detection in early stages could lead to immediately applying appropriate treatment, and implicitly improving the survival rate. The current paper aims to demonstrate that more accurate systems could be designed by combining different convolutional neural networks using different custom combination approaches and by selecting the appropriate networks to be involved in the ensemble model to achieve the best performance metrics. It is essential to understand if combining all experimented networks or only the best-performing ones could always lead to the most effective results or not. The current paper is structured in three main phases. The first step is to propose the individual networks involved in the experiments. Five DeepLab-V3+ networks with different encoders (ResNet-18, ResNet-50, MobileNet-V2, InceptionResNet-V2, and Xception) were used. In the second step, the paper proposes a custom algorithm to combine multiple individual semantic segmentation networks, while the last step describes the iterative selection approach for selecting all individual networks to be combined so that the most accurate ensemble is obtained. The system performing semantic segmentation for different types of ovarian tumors, covering both benign and malignant ones, achieved 91.18 % Intersection over union (IoU), thus overperforming all individual networks. The proposed method could be extended so that more powerful deep learning models could be used.

Semantic Segmentation of Point Cloud Scene via Multi-Scale Feature Aggregation and Adaptive Fusion

Point cloud semantic segmentation is a key step in 3D scene understanding and analysis. In recent years, deep learning–based point cloud semantic segmentation methods have received extensive attention from researchers. Multi-scale neighborhood feature learning methods are suitable for inhomogeneous density point clouds, but different scale branching feature learning increases the computational complexity and makes it difficult to accurately fuse different scale features to express local information. In this study, a point cloud semantic segmentation network based on RandLA-Net with multi-scale local feature aggregation and adaptive fusion is proposed. The designed structure can reduce computational complexity and accurately express local features. The mean intersection-over-union is improved by 1.1% on the SemanticKITTI data set with an inference speed of nine frames per second, while the mean intersection-over-union is improved by 0.9% on the S3DIS data set, compared with RandLA-Net. We also conduct ablation studies to validate the effectiveness of the proposed key structure.

Adults Adapt to Child Speech in Causative Semantics.

Causation is a core feature of human cognition and language. How children learn about intricate causal meanings is yet unresolved. Here, we focus on how children learn verbs that express causation. Such verbs, known as lexical causatives (e.g., break and raise), lack explicit morphosyntactic markers indicating causation, thus requiring that the child generalizes the causal meaning from the context. The language addressed to children presumably plays a crucial role in this learning process. Hence, we tested whether adults adapt their use of lexical causatives to children when talking to them in day-to-day interactions. We analyzed naturalistic longitudinal data from 12 children in the Manchester corpus (spanning from 20 to 36 months of age). To detect semantic generalization, we employed a network approach with semantics learned from cross-situational contexts. Our results show an increasing trend in the expansion of causative semantics, observable in both child speech and child-directed speech. Adults consistently maintain somewhat more intricate causative semantic networks compared to children. However, both groups display evolving patterns. Around 28-30 months of age, children undergo a reduction in the degree of causative generalization, followed by a slightly time-lagged adjustment by adults in their speech directed to children. These findings substantiate adults' adaptation in child-directed speech, extending to semantics. They highlight child-directed speech as a highly adaptive and subconscious teaching tool that facilitates the dynamic processes of languageacquisition.

Hotel-related Research Trends through Semantic Network Analysis: Comparison before and after COVID-19

Hotel-related Research Trends through Semantic Network Analysis: Comparison before and after COVID-19