检索结果-内蒙古大学图书馆

Efficient image Super-Resolution With Feature Interaction Weighted Hybrid Network

ieee TRANSACTIONS ON MULTIMEDIA 2025年 27卷 2256-2267页

作者： Li, Wenjie Li, Juncheng Gao, Guangwei Deng, Weihong Yang, Jian Qi, Guo-Jun Lin, Chia-Wen Beijing Univ Posts & Telecommun Sch Artificial Intelligence Pattern Recognit & Intelligent Syst Lab Beijing 100080 Peoples R China Shanghai Univ Sch Commun & Informat Engn Shanghai 200444 Peoples R China Nanjing Univ Posts & Telecommun Inst Adv Technol IVIPLab Nanjing 210046 Peoples R China Minist Educ Key Lab Artificial Intelligence Shanghai 200240 Peoples R China Soochow Univ Prov Key Lab Comp Informat Proc Technol Suzhou 215006 Peoples R China Nanjing Univ Sci & Technol Sch Comp Sci & Technol Nanjing 210094 Peoples R China Westlake Univ Res Ctr Ind Future Hangzhou 310024 Peoples R China Westlake Univ Sch Engn Hangzhou 310024 Peoples R China OPPO Res Seattle WA 98101 USA Natl Tsing Hua Univ Dept Elect Engn Hsinchu 300044 Taiwan Natl Tsing Hua Univ Inst Commun Engn Hsinchu 300044 Taiwan

Lightweight image super-resolution aims to reconstruct high-resolution images from low-resolution images using low computational costs. However, existing methods result in the loss of middle-layer features due to activation functions. To minimize the impact of intermediate feature loss on reconstruction quality, we propose a Feature Interaction Weighted Hybrid Network (FIWHN), which comprises a series of Wide-residual Distillation Interaction Block (WDIB) as the backbone. Every third WDIB forms a Feature Shuffle Weighted Group (FSWG) by applying mutual information shuffle and fusion. Moreover, to mitigate the negative effects of intermediate feature loss, we introduce Wide Residual Weighting units within WDIB. These units effectively fuse features of varying levels of detail through a Wide-residual Distillation Connection (WRDC) and a Self-Calibrating Fusion (SCF). To compensate for global feature deficiencies, we incorporate a Transformer and explore a novel architecture to combine CNN and Transformer. We show that our FIWHN achieves a favorable balance between performance and efficiency through extensive experiments on low-level and high-level tasks.

关键词： Transformers Convolutional neural networks computational modeling Feature extraction Training Superresolution Adaptation models Computer architecture Telecommunications image reconstruction Single-image super-resolution wide-residual distillation interaction hybrid network transformer

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial Special Issue on Explainable Deep Learning for Medical image processing and Analysis

引用

ieee TRANSACTIONS ON EMERGING TOPICS IN computational intelligence 2023年第1期7卷 2-4页

作者： Zhang, Yu-Dong Gorriz, Juan Manuel Pan, Yi Cordon, Oscar Univ Leicester Leicester LE1 7RH England Univ Cambridge Cambridge CB2 1TN England Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen 518055 Peoples R China Univ Granada Dept Comp Sci Granada 18071 Spain Univ Granada Artificial Intelligence & Andalusian Res Inst Data Granada 18071 Spain

来源：评论

学校读者我要写书评

暂无评论

image processing-Based Methods to Improve the Robustness of Robotic Gripping 22

Image Processing-Based Methods to Improve the Robustness of ...

引用

ieee Joint 22nd International symposium on computational intelligence and Informatics / 8th ieee International Conference on Recent Achievements in Mechatronics, Automation, Computer Science and Robotics (CINTI-MACRo)

作者： Takacs, Kristof Elek, Renata Nagyne Haidegger, Tamas Obuda Univ Antal Bejczy Ctr Intelligent Robot Budapest Hungary Obuda Univ Doctoral Sch Appl Informat & Appl Math Budapest Hungary Obuda Univ John von Neumann Fac Informat Budapest Hungary Obuda Univ Univ Res & Innovat Ctr EKIK Budapest Hungary Austrian Ctr Med Innovat & Technol Wiener Neustadt Austria

ISBN: (纸本)9798350398823

image processing techniques are having a huge impact on most fields of robotics and industrial automation. Real-time methods are usually employed in complex automation tasks, assisting with decision making or directly guiding robots and machinery, while post-processing is usually used for retrospective assessment of systems and processes. While artificial intelligence-based image processing algorithms (relying usually on neural networks) are more common nowadays, "classical" image processing methods can also be used effectively for most modern applications. This paper focuses on optical flow-based image processing, proving its efficiency by presenting optical flow-based solutions for modern challenges in different fields of robotics, such as robot-assisted surgery and food processing. The application domain introduced in this paper is based on a smart robotic gripper designed to support automated robot cells in the meat industry. The gripper is capable of slip detection and secure gripping of soft, slippery tissues with the help of the implemented real-time algorithm.

关键词： image processing Optical flow Slip detection Robotic gripping

来源：评论

学校读者我要写书评

暂无评论

Special Vehicle Classification Algorithm-Based System for Dedicated Parking Zone Violation Detection in South Korea

引用

ieee ACCESS 2025年 13卷 7883-7901页

作者： Park, Hyunseong Kim, Kapyeol Jeong, Incheol Jung, Jungil Cho, Jinsoo Gachon Univ Coll IT Dept IT Convergence Engn Seongnam Si 13120 South Korea PCT Co Ltd Seongnam Si 13449 South Korea Gachon Univ Dept Comp Engn Seongnam Si 13120 South Korea

To address the problem of managing dedicated parking zones arising from the increasing number of electric vehicles and vehicles for the physically challenged, this paper proposes a license plate recognition (LPR)-based parking control system that combines the YOLO and MobileNet algorithms. These two algorithms are designed for real-time object detection and efficient preprocessing, respectively, and can operate in real time in resource-constrained edge-device environments. In tests using data from more than 51,000 vehicles, the system achieved an accuracy rate of 95.76% in classifying electric vehicles and 97.18% in classifying vehicles for the physically challenged. The average CPU and RAM utilizations of the system were 34.54% and 45.04%, respectively. In addition, the processing time per image was recorded as approximately 1.04 s, demonstrating its potential to run reliably on edge devices. These results are expected to facilitate the efficient resolution of parking management problems in smart cities and effective operation of parking zones reserved for electric vehicles and vehicles for the physically challenged.

关键词： Real-time systems YOLO Accuracy License plate recognition Classification algorithms Artificial intelligence Feature extraction computational modeling computational efficiency Smart cities bounding box license plate recognition (LPR) MobileNet optical character recognition (OCR) video processing

来源：评论

学校读者我要写书评

暂无评论

SegBlocks: Block-Based Dynamic Resolution Networks for Real-Time Segmentation

引用

ieee TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE intelligence 2023年第2期45卷 2400-2411页

作者： Verelst, Thomas Tuytelaars, Tinne Katholieke Univ Leuven Ctr Proc Speech & Images Dept Elect Engn B-3000 Leuven Belgium

SegBlocks reduces the computational cost of existing neural networks, by dynamically adjusting the processing resolution of image regions based on their complexity. Our method splits an image into blocks and downsamples blocks of low complexity, reducing the number of operations and memory consumption. A lightweight policy network, selecting the complex regions, is trained using reinforcement learning. In addition, we introduce several modules implemented in CUDA to process images in blocks. Most important, our novel BlockPad module prevents the feature discontinuities at block borders of which existing methods suffer, while keeping memory consumption under control. Our experiments on Cityscapes, Camvid and Mapillary Vistas datasets for semantic segmentation show that dynamically processing images offers a better accuracy versus complexity trade-off compared to static baselines of similar complexity. For instance, our method reduces the number of floating-point operations of SwiftNet-RN18 by 60% and increases the inference speed by 50%, with only 0.3% decrease in mIoU accuracy on Cityscapes.

关键词： image segmentation Complexity theory Task analysis computational efficiency Semantics computational modeling Costs Conditional execution convolutional neural networks model compression semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Neural Architecture Search Framework With Masked Encoding Mechanism for Point Cloud Registration

引用

ieee TRANSACTIONS ON EMERGING TOPICS IN computational intelligence 2025年

作者： Wu, Yue Gong, Peiran Yuan, Yongzhe Gong, Maoguo Ma, Wenping Miao, Qiguang Xidian Univ Sch Comp Sci & Technol Minist Educ Key Lab Collaborat Intelligence Syst Xian 710071 Peoples R China Xidian Univ Sch Elect Engn Minist Educ Key Lab Collaborat Intelligence Syst Xian 710071 Peoples R China Inner Mongolia Normal Univ Acad Artificial Intelligence Coll Math Sci Hohhot 010022 Peoples R China Xidian Univ Sch Artificial Intelligence Minist Educ Key Lab Intelligent Percept & Image Understanding Xian 710071 Peoples R China

Deep learning-based point cloud registration (PCR) has received increasing attention in the field of computer vision. However, the lack of expertise in deep neural network architecture design remains a large obstacle for end-users with registration needs. While neural architecture search (NAS) indeed addresses the architecture automation design problem and has undergone substantial development, the majority of existing NAS approaches are tailored for image processing. Because of significant differences in data characteristics and network structures, directly transferring the existing NAS approaches for PCR is infeasible. Hence, this work proposes an evolutionary neural architecture search framework for PCR, which can fit the different types of PCR networks. The framework, equipped with the designed masked encoding mechanism, can seamlessly integrate with the majority of evolutionary computation-based optimization algorithms, thereby reducing the additional cost of operator design. Based on the above framework, the users can freely select the advanced network components and optimization algorithms depending on their needs. Experimental results on synthetic and real-world datasets, ModelNet40 and KITTI, demonstrate the superiority of the proposed framework in terms of registration precision and architecture complexity.

关键词： Feature extraction Point cloud compression Encoding Computer architecture Data mining Accuracy Three-dimensional displays Decoding Transformers Neural architecture search Point cloud registration evolutionary neural architecture search novel encoding mechanism

来源：评论

学校读者我要写书评

暂无评论

CVIformer: Cross-View Interactive Transformer for Efficient Stereoscopic image Super-Resolution

引用

ieee TRANSACTIONS ON EMERGING TOPICS IN computational intelligence 2025年第2期9卷 1107-1118页

作者： Zhang, Dongyang Liang, Shuang He, Tao Shao, Jie Qin, Ke Univ Elect Sci & Technol China Inst Intelligent Comp Chengdu 611731 Peoples R China Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu Peoples R China

Inspired by the great success of the Transformer in computer vision, some works have started to explore the use of the Transformer for super-resolution (SR). However, with regard to stereoscopic SR, which aims to recover details from input pairs, how to efficiently integrate cross-view interactions into the Transformer architecture is still an ongoing development. Additionally, most existing stereoscopic SR methods only adopt a parallax mechanism in the middle of the network, and another issue is that the feature correlation from different viewpoints inevitably weakens as the network depth increases. To address these issues, we first utilize an efficient residual transformer block (ERTB) as the backbone for long-range intra-view feature extraction. Subsequently, we propose a novel multi-Dconv cross attentive block (MCAB) to enhance the cross-view interactions at the rear part of the Transformer architecture. Notably, the proposed MCAB promotes feature fusion from two viewpoints by employing bidirectional cross-attention, as opposed to an unidirectional flow from left to right or vice versa. This approach results in an efficient cross-view interaction from both branches. By leveraging the advantages of the proposed ERTB and MCAB, we introduce an efficient cross-view interaction Transformer (CVIformer) for stereoscopic SR. This architecture is capable of incorporating long-range intra-view and cross-view information with an acceptable computational overhead. Without excessive complexity, extensive experiments conducted on four public datasets demonstrate that our model achieves state-of-the-art results using only 1.17 million parameters, with approximately a 40% reduction in parameters compared to leading methods like iPASSR.

关键词： Transformers Stereo image processing Feature extraction Superresolution Spatial resolution Cameras Task analysis Stereoscopic image processing lightweight transformer super-resolution (SR)

来源：评论

学校读者我要写书评

暂无评论

GLGFN: Global-Local Grafting Fusion Network for High-Resolution image Deraining

引用

ieee TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第11期34卷 10860-10873页

作者： Yan, Tao Zhu, Xiangjie Chen, Xianglong He, Weijiang Wang, Chenglong Yang, Yang Wang, Yinghui Chang, Xiaojun Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214122 Jiangsu Peoples R China Jiangsu Univ Dept Comp Sci Zhenjiang 212013 Jiangsu Peoples R China Univ Technol Sydney Australian Artificial Intelligence Inst Fac Engn & Informat Technol Sydney NSW 2007 Australia

image deraining is a hot research topic, which aims to remove various rain streaks (raindrops) from rainy images and restore the backgrounds. Though image deraining has been extensively studied in recent years, few methods are able to effectively and efficiently derain real-world highresolution rainy images. In general, existing image deraining methods are restricted by two main factors while processing high-resolution images. First, the computational complexity and memory usage of existing deep learning-based methods are high when it comes to derain high-resolution images. Second, as the image resolution increases, it is difficult to simultaneously extract and aggregate both global and local features for clean rain removal. In this paper, we propose a novel network, called Global-Local Grafting Fusion Network (GLGFN), for deraining real-world high-resolution images. Our GLGFN utilizes a staggered connection structure to achieve deeper sampling depth while maintaining low computational cost. It adopts the Transformer and CNN based encoders (backbones) to extract global and local features, respectively, and then grafts global features into local features to guide the extraction of rain streaks. In addition, for well fusing global and local features, we also propose a Grafting Fusion Module (GFM), which adopts Cross Sparse Attention (CSA) and Selective Kernel Fusion (SK Fusion) to efficiently aggregate global and local features. Extensive experiments conducted on several high-resolution real rainy datasets have demonstrated the effectiveness and efficiency of our proposed GLGFN. We will release our code and dataset.

关键词： High-resolution image real-world image deraining deep learning staggered connection structure grafting fusion

来源：评论

学校读者我要写书评

暂无评论

SSMamba: Superpixel Segmentation With Mamba

引用

ieee SIGNAL processing LETTERS 2025年 32卷 1715-1719页

作者： Jia, Xiaohong Li, Yonghui Jiao, Jianjun Zhao, Yao Xia, Zhiwei Lanzhou Jiaotong Univ Sch Elect & Informat Engn Lanzhou 730070 Peoples R China Beijing Jiaotong Univ Key Lab Big Data & Artiffcial Intelligence Transpo Ministr Educ Beijing 100044 Peoples R China Beijing Jiaotong Univ Inst Informat Sci Beijing 100044 Peoples R China

Deep convolutional networks have achieved remarkable success in superpixel segmentation. However, they only focus on local features ignoring global attributes. The visual Mamba demonstrates an exceptional capability to capture long-range dependencies and offers a lower computational cost compared to the Transformer. Building on this inspiration, we propose a novel superpixel segmentation with Mamba, termed SSMamba. In SSMamba, Mamba is integrated into a global-local architecture, enabling efficient interaction between global attributes and local features to produce high-quality superpixels. The designed activation function further enhances the effectiveness of SSMamba. Extensive experiments on four public datasets demonstrate that SSMamba outperforms existing state-of-the-art methods, achieving competitive average values of ASA = 0.9541, BR = 0.8768, BP = 0.2124, UE = 0.0910, and CO = 0.3698.

关键词： Feature extraction Head Convolutional neural networks image edge detection computational modeling Visualization Transformers Training Data mining computational efficiency Convolutional neural network mamba superpixel segmentation

来源：评论

学校读者我要写书评

暂无评论

Lightweight Deep Neural Network Model With Padding-Free Downsampling

引用

ieee SIGNAL processing LETTERS 2024年 31卷 865-869页

作者： Liu, Dengfeng Guo, Xiaohe Wang, Ning Wu, Qin Jiangnan Univ Sch Artificial Intelligence & Comp Sci Wuxi 214122 Peoples R China Jiangnan Univ Engn Res Ctr Intelligent Technol Healthcare Minist Educ Wuxi 214122 Peoples R China Wuxi Machinery & Electron Higher Profess & Tech Sc Wuxi 214028 Peoples R China

Deep neural networks have achieved impressive performance in image classification tasks. However, due to limitations in hardware resources, including computing units and storage capacity, deploying these networks directly on resource-constrained devices such as mobile and edge devices is challenging. While lightweight network models have made significant advancements, the downsampling stage has received little attention. As the feature map is reused multiple times, the reduction of its size during the downsampling stage not only reduces the computational cost of the downsampling module itself but also lowers the computational burden of subsequent stages. This letter addresses this gap by proposing a padding-free downsampling module that effectively reduces computational costs and can seamlessly integrates into various deep learning models. Furthermore, we introduce a hybrid stem layer to obtain competitive accuracy. Extensive experiments were conducted on CIFAR-100, Stanford Dogs, and imageNet datasets. On the CIFAR-100 dataset, the results show that the proposed module reduces computational costs by approximately 20% and improves inference speed on resource-constrained devices by around 10%.

关键词： Deep learning image classification model architecture design neural network lightweighting

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：