检索结果-内蒙古大学图书馆

Electronic Information Engineering and Computer Technology (EIECT), International Conference on

作者： Yongwen Guo Hongming Chen Juntao Huang Yilei Lin Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province Zhejiang Ocean University Zhoushan China College of Physics and Optoelectronic Engineering Shenzhen University Shenzhen China

Water contamination is a major concern worldwide due to its impact on the global ecology. Accurate forecasting of water quality parameters is crucial for effective water contamination control. However, existing models face challenges in accurately predicting water quality series due to their nonlinearity and non-stationarity. Additionally, further investigation is needed to find out the performance of the Bidirectional Long Short-Term Memory network (BiLSTM) when combined with optimization of signal processing and adjustment of hyperparameters. In this article, we introduce a novel hybrid approach named CEEMDAN-QPSO-BiLSTM, which combines three techniques - Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Quantum behaved Particle Swarm optimization (QPSO), and BiLSTM - for predicting water quality parameters. To evaluate the effectiveness of this hybrid model, we developed four benchmark models, namely LSTM, BiLSTM, PSO-BiLSTM, and QPSO-BiLSTM, and compared their predictive accuracy with that of our hybrid approach. The training and testing of these models utilized actual water quality data from the Chinese National Marine Environmental Monitoring Center’s official system, specifically focusing on Dissolved Oxygen (DO), pH value, and Chemical Oxygen Demand (COD) parameters. The experimental results show that the CEEMDAN-QPSO-BiLSTM surpasses the baseline models, providing better error performance and prediction accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Research on Semantic Segmentation Method for Maritime Navigation Scene Based on Improved DMNet

Research on Semantic Segmentation Method for Maritime Naviga...

引用

Computer Science, Electronic Information Engineering and Intelligent Control Technology (CEI), IEEE International Conference on

作者： Yumeng Wang Zhenbo Bi Wei Rao Diya Zhang Yanrui Wang Yile Pu Zhejiang Ocean University College of Information Engineering ZhouShan ZheJiang China Key Laboratory Of Marine Big Data Mining And Application Zhejiang Ocean University ZhouShan ZheJiang China

ISBN: (数字)9798350372052

ISBN: (纸本)9798350372069

In recent years, semantic segmentation has been continuously developing, but there are still problems such as incomplete image understanding, incomplete contextual information extraction, imbalanced data samples, and uneven object distribution. A semantic segmentation algorithm based on an improved Dynamic Multi scale Network (DMNet) is proposed to address the current issues. An attention mechanism is added to the decoder to make the network pay more attention to the target to be detected; Using online difficult sample mining to improve the problem of sample imbalance in the dataset; Utilize the CrossEntropyLoss+Lovasz SoftmaxLoss joint loss function and the Adam with decoupled weight Decay (AdamW) optimizer to improve network training effectiveness and segmentation accuracy. The algorithm achieves 85.01% mIoU and 90.30% mAcc, effectively optimizing the model structure and improving segmentation accuracy, thus providing a certain technical foundation and application reference for semantic understanding of maritime channel scenes.

关键词： Training Navigation Semantic segmentation Heuristic algorithms Semantics Transportation Information retrieval

来源：评论

学校读者我要写书评

暂无评论

Feature Enhancement Composition Model for Heavy Rain Removal on Single Image

Feature Enhancement Composition Model for Heavy Rain Removal...

引用

2021 International Conference on Computer Graphics, Artificial Intelligence, and data Processing, ICCAID 2021

作者： Xu, Xufeng Wang, Quanqing Pan, Xiaomiao Pan, Qiang Yang, Yueting Wang, Chao School of Information Engineering Zhejiang Ocean University Zhejiang Zhoushan316022 China School of Mathematics and Statistics Beihua University Jilin Jilin132013 China Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province Zhejiang Zhoushan316022 China

ISBN: (数字)9781510652170

ISBN: (纸本)9781510652163

The current single image rain removal method is not ideal in the task of rain and haze removal. In this paper, based on the physical imaging mechanism under heavy rain conditions, a feature enhanced composition model, FEC-Net, is constructed by combining physical model-driven and data-driven methods. The feature maps, which contain rain streaks, atmospheric light, and transmission map information were obtained by three convolution layers. These feature maps are enhanced by the feature enhancement module. Finally, a clear image is obtained according to the formula of the atmospheric scattering model. The selective enhanced processing of key output data greatly improves the processing effect of the network. Extensive experiments show that our FEC-Net achieves better accuracy and visual improvements against state-of-the-art methods. © 2022 SPIE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Identification of Phtato Leaf Disease Based on The TS-Resnet Deep Learing Method

Identification of Phtato Leaf Disease Based on The TS-Resnet...

引用

International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR)

作者： Xiao-Yue Duan Shen-Ming Gu School of information Engineering Zhejiang Ocean University Zhoushan China Key Laboratory of Oceanographic Big Data Mining&Application of Zhejiang Province Zhejiang Ocean University Zhoushan China

At present, the incidence of potato disease is large, the disease is complex, and it is difficult to control, which seriously affects the potato yield and quality. Therefore, in order to realize the potato disease image recognition technology under small sample data, by using attention mechanism, transfer learning and data enhancement technology, the potato leaf disease image is accurately analyzed. This paper proposes a potato leaf disease recognition method based on TS-ResNet(Transform learning and SE) deep learning method. Based on the original ResNet101 model, its full connectivity layer is modified to add transfer learning and attention mechanism. In order to increase the number of potato disease part of the data set, the original graph can be flipped, horizontal flip and other operations to avoid the phenomenon of overfitting in the original model. The results of this study can provide some support for the intelligent research of potato leaf disease identification.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive disentangled target representation for unsupervised domain adaptation in remote sensing segmentation

引用

Engineering applications of Artificial Intelligence 2025年 156卷

作者： Lu, Runuo Dong, Shoubin Jia, Jianxin Wang, Xusheng Liu, Kai Chen, Jinsong Guo, Shanxin Zheng, Xiaorou Guangdong Provincial Key Laboratory of Multimodal Big Data Intelligent Analysis School of Computer Science and Engineering South China University of Technology Guangzhou510641 China Department of Photogrammetry and Remote Sensing Finnish Geospatial Research Institute EspooFI-02150 Finland Center for Geo-Spatial Information Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China Shenzhen Engineering Laboratory of Ocean Environmental Big Data Analysis and Application Shenzhen518055 China

Due to the significant differences between the source and target domains, semantic segmentation models for remote sensing images trained on the source domain often struggle to generalize effectively to new target domains, leading to performance degradation. Most solutions disentangle features into domain-invariant features shared by both domains and domain-specific features unique to each domain, relying primarily on the domain-invariant features for model construction and inference. However, during this process, critical domain-specific features as essential information are often discarded as noise, leading to a reduction in segmentation accuracy. To address these issues, this study proposes an unsupervised domain adaptation, called ChameleonRS. ChameleonRS separates the domain-invariant features and the domain-specific features by disentanglement learning, and introduces a multi-level feature alignment module to enhance the representation of domain-invariant features. More importantly, it aligns predictions based on domain-invariant features with those based on fused invariant and specific features using contrastive learning, effectively capturing target-specific features. This study explores two cross-domain semantic segmentation tasks: (I) Geographic location variation. (II) Geographic location and imaging mode variation. Results show that under setting (I), ChameleonRS outperforms the state-of-the-art method by 4.79% in Intersection over Union (IoU) and 6.51% in the F1 score. Under setting (II), it achieves improvements of 4.46% and 4.82% in IoU and F1 score. Results demonstrate the superiority and adaptability of ChameleonRS under varying geographical and imaging conditions, effectively mitigating the generalization issues caused by distribution differences, and thereby promoting the development and application of remote sensing technologies in practical scenarios. © 2025 Elsevier Ltd

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A survey of ship target feature extraction based on video surveillance

A survey of ship target feature extraction based on video su...

引用

2021 International Conference on Computer, Communication, Control, Automation and Robotics, CCCAR 2021

作者： Bi, Zhenbo Zhang, Shiyou Wu, Yuanhong Wang, Guangwei Qi, Changshong Information Engineering Zhejiang Ocean University Zhejiang Zhoushan316022 China Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province Zhejiang Zhoushan316022 China

Ship target feature extraction plays an important role in ship recognition and tracking based on video surveillance. However, the current feature extraction of ship targets based on video surveillance does not have a complete summary, so this article reviews these methods. This article summarizes the main ship target feature extraction methods used in ship target recognition and tracking in different application environments, including traditional artificial ship feature extraction and automated learning ship feature extraction. It focuses on the core ideas, advantages and disadvantages of the representative methods in each type of problem, and discusses the current problems in the feature extraction of ship targets and the future development of feature extraction technology. © Published under licence by IOP Publishing Ltd.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Fish Object Detection Based on Spatial Bias Pyramid Pooling: Improved BIAS-YOLO

Fish Object Detection Based on Spatial Bias Pyramid Pooling:...

引用

6th International Conference on Intelligent Computing and Signal Processing (ICSP)

作者： Chunhui Zhang Shenming Gu School of Information Engineering Zhejiang Ocean University Zhou Shan China Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province Zhejiang Ocean University Zhou Shan China

Object detection, as an important research direction of artificial intelligence technology, has a wide scope of application in several fields. At the same time, object detection technology shows a trend of multidisciplinary integration, which frees related researchers and some laborers from repetitive labor, thus improving the efficiency and accuracy of their work. Fishery resources play a pivotal role as an important agricultural resource, and the realization of different kinds of fish detection is of great help to the rational use and protection of fishery resources. However, there are some problems in the application of object detection technology to this field, such as the lack of datasets, and the quantity and quality of samples in the data set cannot satisfy the increasingly large neural network models. In this paper, we created a dataset dedicated to fish target detection through various channels such as photography and the web. The dataset contains 16 categories and 1038 images, including common fish (e.g., carp, hairtail, etc.) and also some precious fish such as grouper. We improved the quality of the images to enable the network to learn more accurate information about the characteristics of different fish. Meanwhile, we propose BIAS-YOLO, which improves the original SPP (Spatial Pyramid Pooling) module based on the original YOLOv5 network, and introduced the idea of spatial bias, which can encode the global information and thus improve the network to make full use of the whole image information. On our proposed dataset, our network works better compared to the baseline.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ship Target Detection Algorithm Based on Improved YOLOv5

Ship Target Detection Algorithm Based on Improved YOLOv5

引用

Computer Science, Electronic Information Engineering and Intelligent Control Technology (CEI), IEEE International Conference on

作者： Louyu Zhu Zhenbo Bi School of Information Engineering Zhejiang Ocean University Zhoushan China Department of Physics and Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province Zhejiang Ocean University Zhoushan China

ISBN: (数字)9798350372052

ISBN: (纸本)9798350372069

In view of the low accuracy and low efficiency of traditional ship target detection methods, this paper introduces an improved ship target detection method based on YOLOv5. Firstly, we preprocess the ship target data set, which includes graph denoising, graph enhancement and so on. Secondly, based on the benchmark YOLOv5 object detection algorithm, SimAM attention mechanism module is introduced. Then, a pyramid feature fusion strategy is added to filter conflicting information in airspace to suppress inconsistent features and improve the network's feature fusion capability for targets of different scales. Finally, the trained model is tested to achieve accurate evaluation of ship target detection. Experimental results show that the proposed ship target detection method is compared with YOLOv5s in accuracy rate, recall rate, mAP@0.5 and mAP@0.5: 0.95 increased by 1.9, 3.8, 2.6 and 7.3 percentage points. Respectively, which can meet the requirements of detection speed and obtain better detection accuracy, effectively realizing high-speed and high-precision ship detection.

关键词： YOLO Computer science Noise reduction Filtering algorithms Benchmark testing Information filters Robustness

来源：评论

学校读者我要写书评

暂无评论

Optimal Scale Selection in Multi-Scale Interval-Set Decision Tables

Optimal Scale Selection in Multi-Scale Interval-Set Decision...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Zhen-Huang Xie Wei-Zhi Wu Lei-Xi Wang School of Information Engineering Zhejiang Ocean University Zhoushan Zhejiang China Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province Zhejiang Ocean University Zhoushan Zhejiang China

In traditional interval-set information systems (ISISs), each attribute is single-scale. However, processing and analyzing data at different scales is often necessary for practical applications. This paper introduces the granular information transformation into multi-scale interval-set information systems (MISISs), defines the similarity relation between objects in MISISs, and defines the rough approximation and correlation properties at different scales studied. Then, we propose the multi-scale interval-set decision table (MISDT) concept and discuss the monotonicity of scale variation of the positive region, generalized decision and positive complementarity conditional entropy. Then, we give three criteria for determining the optimal scales and discuss the relationship among the optimal scales of MISDTs. Finally, give heuristic algorithms and examples.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Frequency-domain Transformer-based Low-light Image Enhancement Network

Frequency-domain Transformer-based Low-light Image Enhanceme...

引用

Computer Technology and Information Science (ISCTIS), International Symposium on

作者： Yang Zhou Qihong Ye Hongming Chen Xiaoshuang Wang School of Information Engineering Zhejiang Ocean University Zhoushan China Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province School of Information Engineering Zhejiang Ocean University Zhoushan China

During the process of capturing images under low-light conditions, images often suffer from issues such as low contrast and high noise due to the limitations of illumination. Recently, with the rapid development of transformers, they have shown significant effectiveness in low-light image enhancement tasks. Therefore, we propose a low-light image enhancement network based on the frequency-domain transformer (FDT). This network consists of a main network and a frequency-domain auxiliary network. In the main network, we use the proposed frequency-domain transformer (FDT) to distinguish between high and low-frequency domains and selectively preserve the required high and low-frequency information. In the frequency-domain auxiliary network, the frequency-domain enhancement module (FDEM) is used to extract features and aggregate with the main network, to achieve the purpose of assisting the integration of information from the main network. Extensive experimental results demonstrate that our proposed method outperforms existing methods to a large extent.

关键词：

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：