检索结果-内蒙古大学图书馆

IAENG International Journal of computer science 2025年第1期52卷 23-31页

作者： Guo, Ying Wang, Li School of Computer Science and software Engineering University of Science and Technology Liaoning Anshan114051 China College of Computer Science and Technology Liaoning Anshan114051 China

End-to-end training has emerged as a prominent trend in speech recognition, with Conformer models effectively integrating Transformer and CNN architectures. However, their complexity and high computational cost pose deployment challenges. To address these issues, we propose a multi-task Chinese speech recognition method based on the Squeezeformer model. We replace the FMCF structure in Conformer with an MF/CF structure, leveraging the convolutional module as a local Multi- Head Attention (MHA) module to enhance efficiency. Multilevel down-sampling and up-sampling using a time-series U-Net further reduce computational costs. By eliminating redundant LayerNorm layers and employing depthwise separable convolutions, we streamline the model, reduce parameters, and lower deployment costs. An Adaptor Layer is integrated into the MHSA module to mitigate the vanishing gradient problem, and a ScaleVar Layer is added to enhance flexibility. Additionally, the RealFormer module is introduced on the decoding side to improve context understanding. Combining Connectionist Temporal Classification (CTC) with attention-based encoding and decoding models for multi-task learning improves performance and accuracy. Experimental results show that the proposed method reduces the parameters on AISHELL-1 dataset by 16% and reduces the character error rate to 5.50%. At the same time, it also shows good performance on AISHELL-2 dataset. © (2025), (International Association of Engineers). All rights reserved.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Object Detection Model for Remote Sensing Images Based on YOLOv9

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2025年第3期52卷 840-847页

作者： Hou, Donghao Zhang, Yujun School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China

In the field of object detection for remote sensing images, especially in applications such as environmental monitoring and urban planning, significant progress has been made. This paper addresses the common challenges faced by traditional object detection methods in remote sensing images, such as the large number of targets and complex backgrounds, by proposing a novel network based on YOLOv9. The network innovatively introduces the C3_CD_CGA module, an enhanced module based on Cascaded Group Attention, designed to reduce computational redundancy and increase attention diversity, and enhances the processing capability of multi-scale information through the CD module. The C3 module employs deep asymmetric convolution to mitigate information loss and increase the receptive field. Additionally, the network integrates DSConv with the RepNCSPELAN4 module to adaptively focus on and precisely capture the features of elongated and curved local structures, such as vehicles. The introduction of the CARAFE module further improves the spatial resolution of the feature maps, significantly enhancing performance across various visual tasks. Experimental results show that the improved YOLOv9 achieves a mean average precision (mAP) of 88% on the SIMD dataset, which is an improvement of 1.6% compared to the baseline YOLOv9 model and 1.5% higher than the state-of-the-art YOLO-SE model. This model not only achieves more effective multi-target recognition in complex backgrounds but also strikes a good balance between accuracy and efficiency. © (2025), (International Association of Engineers). All rights reserved.

关键词： Urban planning

来源：评论

学校读者我要写书评

暂无评论

Analysis of recent techniques in marine object detection: a review

引用

Multimedia Tools and Applications 2024年 1-99页

作者： Bhalla, Shikha Kumar, Ashish Kushwaha, Riti School of Computer Science Engineering and Technology Bennett University Greater Noida India

Marine Object Detection, leveraging computer vision, plays a vital role in detecting objects in marine environments ranging from marine organisms to marine surveillance. However, marine environment presents unique set of challenges that makes the detection process tedious. Marine environment often has low visibility due to color cast, haziness and intricate background. Moreover, marine target objects appear at varying scales and occluded by the surrounding objects. For comprehensive understanding of this domain, we have analyzed the significant studies from the year (2012-2024) in this work. Prisma model is utilized to identify the crucial 111 studies for this review. The review organizes marine object detection techniques into conventional and advance learning approach through the proposed hierarchical tree-like structure. The organized research has been tabulated, offering an in-depth summary of significant work and highlighting the strengths and weakness of the identified studies for the conventional and advance learning approach. Furthermore, a brief examination of 31 benchmark datasets categorized according to the different application domains (marine species, ship and debris) has been conducted and their key attributes are tabulated. In addition, statistical analysis has been performed to understand the growth and future directions in marine object detection domain. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

HA-FGOVD: Highlighting Fine-Grained Attributes via Explicit Linear Composition for Open-Vocabulary Object Detection

引用

IEEE Transactions on Multimedia 2025年 27卷 3171-3183页

作者： Ma, Yuqi Liu, Mengyin Zhu, Chao Yin, Xu-Cheng University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China

Open-vocabulary object detection (OVD) models are considered to be Large Multi-modal Models (LMM), due to their extensive training data and a large number of parameters. Mainstream OVD models prioritize object coarse-grained category rather than focus on their fine-grained attributes, e.g., colors or materials, thus failed to identify objects specified with certain attributes. Despite being pretrained on large-scale image-text pairs with rich attribute information, their latent feature space does not highlight these fine-grained attributes. In this paper, we introduce HA-FGOVD, a universal and explicit method that enhances the attribute-level detection capabilities of frozen OVD models by highlighting fine-grained attributes in explicit linear space. Our approach uses a LLM to extract attribute words in input text as a zero-shot task. Then, token attention masks are adjusted to guide text encoders in extracting both global and attribute-specific features, which are explicitly composited as two vectors in linear space to form a new attribute-highlighted feature for detection tasks. The composition weight scalars can be learned or transferred across different OVD models, showcasing the universality of our method. Experimental results show that HA-FGOVD achieves state-of-the-art performance on the FG-OVD benchmark and demonstrates promising generalization on the OVDEval benchmark, suggesting that our method addresses significant limitations in fine-grained attribute detection and has potential for broader fine-grained detection applications. © 1999-2012 IEEE.

关键词： Coarse-grained modeling

来源：评论

学校读者我要写书评

暂无评论

UCD_Net: dilated convolution-enhanced upsampling fusion for advanced lung disease classification

引用

Multimedia Tools and Applications 2025年 1-19页

作者： Sharma, Suchit Shelke, Nitin Arvind School of Computer Science Engineering and Technology Bennett University Greater Noida201310 India

Deep Learning has recently been in trend when it comes to medical image analysis as it uses Convolution Neural Network (CNN), which utilizes multi-layer processing to extract intricate and complex features from the data. Similar to other research work, we used Chest X-ray (CXR) images as our dataset and classified them into five categories of lung disease, namely bacterial pneumonia, COVID-19, normal, pneumothorax, and tuberculosis. This study proposes a novel CNN, UCD_Net, which uses a convolutional upsampling layer featuring bicubic interpolation along with a stack of convolution layers before inputting the images into the fine-tuned Xception model and resorts to multiple dilated convolution layers before predicting output. This 2D CNN model is used to achieve optimal feature extraction within minimal time and space constraints. The dataset was divided into a ratio of 0.66 and 0.33 for the training set and testing set, respectively. The best performance is recorded by our proposed model, UCD_Net, with a training accuracy equal to 99.23%, a validation accuracy equal to 98.05%, a recall of 97.8%, a precision of 97.2%, and a fi-score equal to 97.6%. The model performance on the same dataset is also compared to four state-of-the-art architectures, namely ResNet 50, VGG16, Xception, and ResNet152. After our UCD_Net, the best performance was seen using ResNet 50, with accuracy, recall, and precision of 95.45%, 94.5%, and 95.2%, respectively. Training the model on a larger dataset and subsequent fine-tuning can lead to more research and model development. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Individualized Driving Intention Prediction With Inverse Reinforcement Learning

引用

IEEE Transactions on Intelligent Transportation Systems 2025年第6期26卷 8125-8139页

作者： Liu, Siqi Li, Xinyang Chen, Jiansheng Guo, Chenghao Wu, Jiehui Luo, Qifeng Ma, Huimin University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China

Advanced Driver Assistance Systems (ADAS) are designed to prevent collisions, identify the condition of drivers while operating vehicles, and provide additional information to enhance drivers' awareness of potential hazards on the road. Today, ADAS are capable of predicting drivers' actions several seconds in advance, preparing for potential future hazards to prevent accidents or reduce injuries to occupants. Most previous works have achieved prediction results by analyzing and processing a vast amount of driving data from multiple drivers, based on the macro intention preferences of multiple drivers and external environmental features, collectively referred to as the generalized intention prediction network. This network utilizes extensive driving data to predict the common driving intentions of the overall driving population, without considering individualized driving styles. However, according to our research, different drivers exhibit distinct latent preferences in real-world driving scenarios. The generalized intention prediction network is influenced by these latent preferences, resulting in poor generalization capabilities and inaccurate predictions across different drivers. In this study, we propose a individualized driver intention prediction network. Based on Inverse Reinforcement Learning (IRL), it extracts individualized driving intention feature preferences that influence driving intentions from the driver's historical behavior to improve generalized prediction results and achieve individualized driving intention prediction. We demonstrate that preferences vary among different drivers in the driving domain, leading to biases in model predictions. Upon experimental validation, the method we have proposed demonstrates remarkable efficacy on both the Brain4Cars and IESDD datasets, thereby showcasing its enhanced applicability in real-world scenarios. © 2000-2011 IEEE.

关键词： Advanced driver assistance systems

来源：评论

学校读者我要写书评

暂无评论

Preface:Energy and sustainable development

引用

Journal of Central South university 2023年第11期30卷 3517-3518页

作者： Majid SIAVASHI Müslüm ARICI School of Mechanical Engineering Iran University of Science and TechnologyTehranIran Department of Mechanical Engineering Kocaeli UniversityKocaeli41001Turkey

Energy demands are increasing continuously,and the development of sustainable technologies in energy systems is a critical and vital *** the required energy for social and economic needs in line with diminishing the environmental effects and pollution emissions is necessary to meet sustainable *** new energy systems with improved efficiency,securing the energy need,using clean and renewable energy sources,reducing energy losses and pollution,and expanding the availability and access to energy are the key elements of sustainable energy development.

关键词： pollution sustainable curing

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Object Detection Algorithm for Remote Sensing Images

引用

engineering Letters 2025年第3期33卷 704-711页

作者： Hou, Donghao Zhang, Yujun Ren, Jia School of Computer and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Electronic and Information Engineering University of Science and Technology Liaoning Anshan114051 China

With the continuous advancement of satellite technology, remote sensing images has been increasingly applied in fields such as urban planning, environmental monitoring, and disaster response. However, remote sensing images often feature small target sizes and complex backgrounds, posing significant computational challenges for object detection tasks. To address this issue, this paper proposes a lightweight remote sensing images object detection algorithm based on YOLOv9. The proposed algorithm incorporates the SimRMB module, which effectively reduces computational complexity while improving the efficiency and accuracy of feature extraction. Through a dynamic attention mechanism, SimRMB is capable of focusing on important regions while minimizing background interference, and by integrating residual learning and skip connections, it ensures the stability of deep networks. To further enhance detection performance, the FasterRepNCSPELAN4 module is introduced, which employs PConv operations to reduce computational load and memory usage. It also utilizes dilated convolutions and DFC attention mechanisms to strengthen feature extraction, thereby increasing the efficiency and accuracy of object detection. Additionally, this study integrates the GhostModuleV2 module, which generates core feature maps and employs lightweight operations to create redundant features, greatly reducing the computational complexity of *** results show that on the SIMD dataset, the improved YOLOv9 model has a parameter size of 167.88 MB and GFLOPs of 208.6. Compared to the baseline YOLOv9 model (parameter size: 194.57 MB, GFLOPs: 239.0), the parameter size is reduced by 13.71%, GFLOPs are reduced by 12.72%, and detection accuracy is improved by 1.4%. These results demonstrate that the proposed lightweight YOLOv9 model effectively reduces computational overhead while maintaining excellent detection performance, providing an efficient solution for object detection tasks in resou

关键词： Satellite imagery

来源：评论

学校读者我要写书评

暂无评论

Drug-Target Interactions Prediction Based on Signed Heterogeneous Graph Neural Networks

引用

Chinese Journal of Electronics 2024年第1期33卷 231-244页

作者： Ming CHEN Yajian JIANG Xiujuan LEI Yi PAN Chunyan JI Wei JIANG College of Information Science and Engineering Hunan Normal University School of Computer Science Shaanxi Normal University Faculty of Computer Science and Control Engineering Shenzhen Institute of Advanced TechnologyChinese Academy of Sciences Computer Science Department BNU-HKBU United International College

Drug-target interactions(DTIs) prediction plays an important role in the process of drug *** computational methods treat it as a binary prediction problem, determining whether there are connections between drugs and targets while ignoring relational types information. Considering the positive or negative effects of DTIs will facilitate the study on comprehensive mechanisms of multiple drugs on a common target, in this work, we model DTIs on signed heterogeneous networks, through categorizing interaction patterns of DTIs and additionally extracting interactions within drug pairs and target protein pairs. We propose signed heterogeneous graph neural networks(SHGNNs), further put forward an end-to-end framework for signed DTIs prediction, called SHGNN-DTI,which not only adapts to signed bipartite networks, but also could naturally incorporate auxiliary information from drug-drug interactions(DDIs) and protein-protein interactions(PPIs). For the framework, we solve the message passing and aggregation problem on signed DTI networks, and consider different training modes on the whole networks consisting of DTIs, DDIs and PPIs. Experiments are conducted on two datasets extracted from Drug Bank and related databases, under different settings of initial inputs, embedding dimensions and training modes. The prediction results show excellent performance in terms of metric indicators, and the feasibility is further verified by the case study with two drugs on breast cancer.

关键词： Drug-target interactions Signed heterogeneous network Link sign prediction Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Liquid lens based holographic camera for real 3D scene hologram acquisition using end-to-end physical model-driven network

引用

Light(science & Applications) 2024年第3期13卷 488-497页

作者： Di Wang Zhao-Song Li Yi Zheng You-Ran Zhao Chao Liu Jin-Bo Xu Yi-Wei Zheng Qian Huang Chen-Liang Chang Da-Wei Zhang Song-Lin Zhuang Qiong-Hua Wang School of Instrumentation and Optoelectronic Engineering Beihang UniversityBeijing 100191China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China

With the development of artificial intelligence,neural network provides unique opportunities for holography,such as high fidelity and dynamic *** to obtain real 3D scene and generate high fidelity hologram in real time is an urgent ***,we propose a liquid lens based holographic camera for real 3D scene hologram acquisition using an end-to-end physical model-driven network(EEPMD-Net).As the core component of the liquid camera,the first 10 mm large aperture electrowetting-based liquid lens is proposed by using specially fabricated *** design of the liquid camera ensures that the multi-layers of the real 3D scene can be obtained quickly and with great imaging *** EEPMD-Net takes the information of real 3D scene as the input,and uses two new structures of encoder and decoder networks to realize low-noise phase *** comparing the intensity information between the reconstructed image after depth fusion and the target scene,the composite loss function is constructed for phase optimization,and the high-fidelity training of hologram with true depth of the 3D scene is realized for the first *** holographic camera achieves the high-fidelity and fast generation of the hologram of the real 3D scene,and the reconstructed experiment proves that the holographic image has the advantage of low *** proposed holographic camera is unique and can be used in 3D display,measurement,encryption and other fields.

关键词： calculation solution holographic

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：