检索结果-内蒙古大学图书馆

2025 ieee symposium on computational intelligence in Natural Language processing and Social Media, CI-NLPSoMe 2025 2025年

作者： Mizuta, Takanobu

来源：评论

学校读者我要写书评

暂无评论

MANTIS: A Mixed-Signal Near-Sensor Convolutional imager SoC Using Charge-Domain 4b-Weighted 5-to-84-TOPS/W MAC Operations for Feature Extraction and Region-of-Interest Detection

引用

ieee JOURNAL OF SOLID-STATE CIRCUITS 2025年第3期60卷 934-948页

作者： Lefebvre, Martin Bol, David Catholic Univ Louvain Inst Informat & Commun Technol Elect & Appl Math B-1348 Louvain La Neuve Belgium Delft Univ Technol Dept Microelect NL-2628 CD Delft Netherlands

Recent advances in artificial intelligence (AI) have prompted the search for enhanced algorithms and hardware to support the deployment of machine learning (ML) at the edge. More specifically, in the context of the Internet of Things (IoT), vision chips must be able to fulfill the tasks of low to medium complexity, such as feature extraction (FE) or region-of-interest (RoI) detection, with a sub-mW power budget imposed by the use of small batteries or energy harvesting. Mixed-signal vision chips relying on in-or near-sensor processing have emerged as an interesting candidate because of their favorable tradeoff between energy efficiency (EE) and computational accuracy compared with digital systems for these specific tasks. In this article, we introduce a mixed-signal convolutional imager system-on-chip (SoC) codenamed MANTIS, featuring a unique combination of large 16 x 16 4b-weighted filters, operation at multiple scales, and double sampling, well suited to the requirements of medium-complexity tasks. The main contributions are (i) circuits called DS3 units combining delta-reset sampling (DRS), image downsampling (DS), and voltage downshifting and (ii) charge-domain multiply-and-accumulate (MAC) operations based on switched-capacitor (SC) amplifiers and charge sharing in the capacitive DAC of the successive-approximation (SAR) ADCs, MANTIS achieves peak EEs normalized to 1b operations of 4.6 and 84.1 TOPS/W at the accelerator and SoC levels, while computing feature maps (fmaps) with a root-mean-square error (RMSE) ranging from 3 to 11.3%. It also demonstrates a face RoI detection with a false negative rate (FNR) of 11.5%, while discarding 81.3% of image patches and reducing the data transmitted off chip by 13 x compared with the raw image.

关键词： Charge domain CMOS image sensor (CIS) convolutional neural network (CNN) feature extraction (FE) mixed signal multiply-and-accumulate (MAC) operations near sensor region-of-interest (RoI) detection system-on-chip (SoC) Charge domain CMOS image sensor (CIS) convolutional neural network (CNN) feature extraction (FE) mixed signal multiply-and-accumulate (MAC) operations near sensor region-of-interest (RoI) detection system-on-chip (SoC)

来源：评论

学校读者我要写书评

暂无评论

DE-Unet: Dual-encoder U-Net for Ultra-high Resolution Remote Sensing image Segmentation

引用

ieee Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12290-12302页

作者： Liu, Ye Song, Shitao Wang, Miaohui Gao, Hao Liu, Jun Nanjing University of Posts and Telecommunications School of Automation and Artificial Intelligence China Shenzhen University Guangdong Key Laboratory of Intelligent Information Processing Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China Lancaster University School of Computing and Communications United Kingdom

In recent years, there has been a growing demand for remote sensing image semantic segmentation in various applications. The key to semantic segmentation lies in the ability to globally comprehend the input image. While recent transformer-based methods can effectively capture global contextual information, they suffer from high computational complexity, particularly when it comes to ultra-high resolution (UHR) remote sensing images, it is even more challenging for these methods to achieve a satisfactory balance between accuracy and computation speed. To address these issues, we propose in this paper a CNN-based dual-encoder U-Net for effective and efficient UHR image segmentation. Our method incorporates dual encoders into the symmetrical framework of U-Net. The dual encoders endow the network with strong global and local perception capabilities simultaneously, while the U-Net's symmetrical structure guarantees the network's robust decoding ability. Additionally, multi-path skip connections ensure ample information exchange between the dual encoders, as well as between the encoders and decoders. Furthermore, we proposes a context-aware modulation fusion module that guides the encoder-encoder and encoder-decoder data fusion through global receptive fields. Experiments conducted on public UHR remote sensing datasets such as the Inria Aerial and DeepGlobe have demonstrated the effectiveness of proposed method. Specifically on the Inria Aerial dataset, our method achieves a 77.42% mIoU which outperforms the baseline [1] by 3.14% while maintaining comparable inference speed. © 2008-2012 ieee.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

ieee symposium on CI in Health and Medicine (ieee CIHM 2025)

2025 IEEE Symposium on Computational Intelligence in Natural...

引用

2025 ieee symposium on computational intelligence in Natural Language processing and Social Media, CI-NLPSoMe Companion 2025 2025年

作者： Plagianakos, Vassilis P.

来源：评论

学校读者我要写书评

暂无评论

HiFiSeg: High-Frequency Information Enhanced Polyp Segmentation With Global-Local Vision Transformer

引用

ieee ACCESS 2025年 13卷 38704-38713页

作者： Ren, Jingjing Zhang, Xiaoyong Zhang, Lina Taiyuan Univ Dept Intelligence & Informat Engn Taiyuan 030032 Peoples R China

Numerous studies have demonstrated the strong performance of Vision Transformer (ViT)-based methods across various computer vision tasks. However, ViT models often struggle to effectively capture high-frequency components in images, which are crucial for detecting small targets and preserving edge details, especially in complex scenarios. This limitation is particularly challenging in colon polyp segmentation, where polyps exhibit significant variability in structure, texture, and shape. High-frequency information, such as boundary details, is essential for achieving precise semantic segmentation in this context. To address these challenges, we propose HiFiSeg, a novel network for colon polyp segmentation that enhances high-frequency information processing through a global-local vision transformer framework. HiFiSeg leverages the pyramid vision transformer (PVT) as its encoder and introduces two key modules: the global-local interaction module (GLIM) and the selective aggregation module (SAM). GLIM employs a parallel structure to fuse global and local information at multiple scales, effectively capturing fine-grained features. SAM selectively integrates boundary details from low-level features with semantic information from high-level features, significantly improving the model's ability to accurately detect and segment polyps. Extensive experiments on five widely recognized benchmark datasets demonstrate the effectiveness of HiFiSeg for polyp segmentation. Notably, the mDice scores on the challenging CVC-ColonDB and ETIS datasets reached 0.826 and 0.822, respectively, underscoring the superior performance of HiFiSeg in handling the specific complexities of this task.

关键词： Transformers Feature extraction image segmentation Computer vision Convolution Semantics computational modeling Biomedical imaging Deep learning Decoding Polyp segmentation colonoscopy deep learning pyramid vision transformer

来源：评论

学校读者我要写书评

暂无评论

Advancing Natural Language to SQL: A Comparative Study of Open Source LLMs on Benchmark Datasets

Advancing Natural Language to SQL: A Comparative Study of Op...

引用

2025 ieee symposium on computational intelligence in Natural Language processing and Social Media, CI-NLPSoMe Companion 2025

作者： Sharma, Ashwin K Kanumuri, Sri Charan Pranav, A.R. Sharath, Shylaja S Pes University Department of Cse Bengaluru India

ISBN: (纸本)9798331519742

Translating Natural Language to SQL (NL-to-SQL) allows users to communicate with databases using a common language instead of complicated query syntax. This is important as it frees the non-technical user from querying the database, making interactions more easier. This work primarily evaluates open-source language models on three key NL-to-SQL bench-mark datasets: SPIDER, a cross-domain dataset;WikiSQL, a simpler single-domain dataset;and BIRD, a dataset comprising of real-world data. These datasets were selected because of the variety of problems they present: WikiSQL concentrates on effective querying using straightforward single-tables, while BIRD simulates noisy, real-world data scenarios where precise and effective SQL creation is essential. SPIDER analyzes how well models generalize over complex schemas and multi-table queries. We fine-tuned eight models from the LLM families of LLama, Gemma, and Mistral across these datasets with consistent hyper-parameters to tackle these challenges. We used an SQL evaluation framework that examines model generalization, considering both linguistic variety and complex query patterns against a baseline of accuracy and correctness of existing NL-to-SQL models available. This study demonstrates how different models perform across datasets, with some performing better in simpler queries and others being suited for managing inconsistent, real-world data or difficult cross-domain tasks. © 2025 ieee.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Cross-Modal Hashing via Diverse Instances Matching

引用

ieee TRANSACTIONS ON image processing 2025年 34卷 2737-2749页

作者： Tu, Junfeng Liu, Xueliang Huang, Zhen Hao, Yanbin Hong, Richang Wang, Meng Hefei Univ Technol Sch Comp Sci & Informat Engn Hefei 230009 Peoples R China Inst Artificial Intelligence Hefei Comprehens Natl Sci Ctr Hefei 230088 Peoples R China Natl Univ Def Technol Coll Comp Changsha 430074 Peoples R China

Cross-modal hashing is a highly effective technique for searching relevant data across different modalities, owing to its low storage costs and fast similarity retrieval capability. While significant progress has been achieved in this area, prior investigations predominantly concentrate on a one-to-one feature alignment approach, where a singular feature is derived for similarity retrieval. However, the singular feature in these methods fails to adequately capture the varied multi-instance information inherent in the original data across disparate modalities. Consequently, the conventional one-to-one methodology is plagued by a semantic mismatch issue, as the rigid one-to-one alignment inhibits effective multi-instance matching. To address this issue, we propose a novel Diverse Instances Matching for Cross-modal Hashing (DIMCH), which explores the relevance between multiple instances in different modalities using a multi-instance learning algorithm. Specifically, we design a novel diverse instances learning module to extract a multi-feature set, which enables our model to capture detailed multi-instance semantics. To evaluate the similarity between two multi-feature sets, we adopt the smooth chamfer distance function, which enables our model to incorporate the conventional similarity retrieval structure. Moreover, to sufficiently exploit the supervised information from the semantic label, we adopt the weight cosine triplet loss as the objective function, which incorporates the multilevel similarity among the multi-labels into the training procedure and enables the model to mine the multi-label correlation effectively. Extensive experiments demonstrate that our diverse hashing embedding method achieves state-of-the-art performance in supervised cross-modal hashing retrieval tasks.

关键词： Feature extraction Semantics Transformers Data mining Cross modal retrieval Correlation Codes Learning systems computational modeling Cellular phones Cross-modal hashing hashing multi-instance learning

来源：评论

学校读者我要写书评

暂无评论

Real-World image Reflection Removal: An Ultra-High-Definition Dataset and an Efficient Baseline

引用

ieee TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2025年第5期35卷 4397-4408页

作者： Chen, Guang-Yong Zheng, Chao-Wei Fan, Guodong Su, Jian-Nan Gan, Min Philip Chen, C. L. Fuzhou Univ Coll Comp & Data Sci Fuzhou 350116 Peoples R China Minist Educ Fujian Key Lab Network Comp & Intelligent Informat Key Lab Intelligent Metro Univ Fujian Fuzhou 350108 Peoples R China Qingdao Univ Coll Comp Sci & Technol Qingdao 266071 Peoples R China Putian Univ New Engn Ind Coll Putian 351100 Fujian Peoples R China Putian Univ Putian Elect Informat Ind Technol Res Inst Putian 351100 Fujian Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510641 Peoples R China Minist Educ Engn Res Ctr Big Data Intelligence Fuzhou 350108 Peoples R China

Reflection removal is a crucial issue in image reconstruction, especially for high-definition images. Removing undesirable reflections can greatly enhance the performance of various visual systems, such as medical imaging, autonomous driving, and security surveillance. However, the resolution of existing reflection removal datasets is not high and the training data heavily relies on synthetic data, which hampers the performance of reflection removal methods and restricts the development of effective techniques tailored for high-definition images. Therefore, this paper introduces a new dataset, Real-world Reflection Removal in 4K (RR4K). This novel dataset, with its large capacity and high resolution of 6000x 4000 pixels, represents a significant advancement in the field, ensuring a realistic and high quality benchmark. Furthermore, building upon the dataset, we propose an efficient method for single-image reflection removal, optimized for high-definition processing. This method employs the U-Net architecture, enhanced with large kernel distillation and scale-aware features, enabling it to effectively handle complex reflection scenarios while reducing computational demands. Comprehensive testing on the RR4K dataset and existing low-resolution datasets has demonstrated the method's superior efficiency and effectiveness. We believe that our constructed RR4K dataset can better evaluate and design algorithms for removing undesirable reflection from real-world high-definition images. Our dataset and code are available at https://***/jengchauwei/RR4K.

关键词： Reflection image resolution Feature extraction Photography Kernel image reconstruction Glass Deep learning Circuits and systems Benchmark testing Single-image reflection removal image reconstruction benchmark dataset

来源：评论

学校读者我要写书评

暂无评论

A Two-Stage Framework With Ore-Detect and Segment Anything Model for Ore Particle Segmentation and Size Measurement

引用

ieee SENSORS JOURNAL 2025年第7期25卷 11722-11736页

作者： Li, Fei Liu, Xiaoyan Li, Zongping Hunan Univ Coll Elect & Informat Engn Natl Engn Lab Robot Visual Percept & Control Techn Changsha 410082 Peoples R China Hunan Univ Innovat Inst Ind Design & Machine Intelligence Qua Quanzhou 362006 Peoples R China Zhongye Changtian Natl Sintered Pelletizing Engn T Changsha 410082 Peoples R China

This article proposes a novel detection-driven method for ore particle (OP) segmentation, specifically designed to enable precise measurement of particle size distribution (PSD), a critical factor in mineral processing. Accurate segmentation is essential for calculating particle size, which directly contributes to the evaluation of PSD and ensures effective ore quality control and processing efficiency. Conventional segmentation methods often struggle to generalize across diverse ore types, irregular particle geometries, and complex spatial arrangements, limiting their robustness and adaptability in real-world applications. The proposed method overcomes these challenges by first using Ore-Detect, a detection module that efficiently localizes OPs, followed by a refined segmentation process using the segment anything model (SAM) to precisely delineate particle boundaries. This two-stage approach ensures high accuracy in particle segmentation and size measurement, even in the presence of occlusions and significant size variations. Experimental results demonstrate the effectiveness of the proposed method, achieving state-of-the-art performance across key metrics: mean intersection over union (mIoU) of 85.31%, precision of 88.81%, recall of 87.12%, F1 -score of 87.95%, average precision (AP) of 94.06%, boundary displacement error (BDE) of 6.98, and object-level accuracy (OLA) of 89.51%. Furthermore, the framework achieves a PSD measurement error within 5%. In terms of computational efficiency, the proposed method processes 100 ore images of $512\times 512$ resolution in 11.27 s, demonstrating its ability to meet real-time industrial requirements.

关键词： image segmentation Ores Accuracy Particle measurements Feature extraction Atmospheric measurements Shape Real-time systems Size measurement Annotations Detection driven ore particle (OP) segmentation Ore-Detect particle size distribution (PSD) segment anything model (SAM)

来源：评论

学校读者我要写书评

暂无评论

A Modified Parallel NET (MPNET)-Based Deep Learning Technique for the Segmentation and Quantification of Visceral and Superficial Adipose Tissues of CT Scans

引用

ieee ACCESS 2025年 13卷 28522-28538页

作者： Adekanbi, Josteve Das, Debashish Elmitwally, Nouh Ali, Aliyuda Madai, Vince I. Bhatia, Bahadar Morlese, John Afsal, Muhammed Shah, Tanay Birmingham City Univ Fac Comp Engn & Built Environm Birmingham B4 7XG England Sandwell & West Birmingham Hosp NHS Trust West Bromwich B71 4HJ England Cairo Univ Fac Comp & Artificial Intelligence Giza 12613 Egypt Sheffield Hallam Univ Coll Business Technol & Engn Dept Comp Sheffield S1 1WB England Univ Med Berlin Charite Berlin Inst Hlth BIH Charite QUEST Ctr Responsible Res D-10117 Berlin Germany Univ Leicester Space Res Ctr Phys & Astron Leicester LE1 7RH Leics England

This study introduces a modified parallel net (MPNET), a novel deep learning model designed for accurate segmentation and quantification of visceral and superficial adipose tissues. This was used to quantify the visceral and superficial adipose tissues found at the L3 levels of vertebra in CT scans. This will be used to predict the likelihood of the patient developing diabetes or cardiovascular diseases from existing CT scan data. MPNET was compared with state-of-the-art models like UNET, R2UNET, UNET++, and nnUNET. This approach advances the accuracy and efficiency of image segmentation demonstrating a faster learning curve and lower losses at early epochs than traditional models., We developed and validated using a limited dataset of 14 single-slice DICOM files for each patient extracted from the National Health Service UK. The outputs from MPNET not only matched but often exceeded traditional metrics such as the Dice coefficient and IoU in nuanced anatomical delineation, providing greater clinical realism and applicability in segmentation results. As a pilot study, this research paves the way for a forthcoming validation study on a larger and more ethnically diverse dataset.

关键词： image segmentation Computed tomography Accuracy Deep learning DICOM Training Transformers Analytical models Measurement Manuals Convolutional neural networks CT scan analysis deep learning image post-processing segmentation and quantification visceral and superficial adipose tissues

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：