检索结果-内蒙古大学图书馆

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

作者： Liquan Wang Shu Zhan School of Computer and Information Engineering Hefei University of Technology Key Laboratory of Big Data Knowledge Engineering Ministry of Education Hefei China

Micro Expression (ME) is the subtle facial expressions that people show when they express their inner feelings. To address the problem that micro-expression recognition is difficult and less accurate due to the small number of samples and uneven distribution of different categories, we propose a model framework to improve the accuracy of micro-expression recognition. The peak frames containing more key expression information in the micro-expression video sequences are extracted; SE-ResNeXt-50, an improved residual network with SE module, is used to extract features from the peak frames of micro-expressions, where the SE module can better learn the key information in the features, and ResNeXt simplifies the structure by replacing the dense structure with the sparse structure through group convolution, which improves the recognition efficiency. The recognition efficiency is improved by replacing the dense structure with the sparse structure by group convolution. At the same time, the Focal Loss loss function can better solve the model performance problem caused by the imbalance of micro-expression data. Simulation experiments are conducted on the micro-expression dataset CASMEⅡ, and it is found that the improved residual network and peak frame improve the accuracy and F1 value of micro-expression recognition. The improved residual network and peak frame can reduce the effect of small data set, make the model have good fitting effect, and improve the performance of different categories, improve the recognition accuracy of micro-expressions, and have better recognition performance for micro-expression recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fast Multivariate Time Series Anomaly Detection Based on Matrix Completion

Fast Multivariate Time Series Anomaly Detection Based on Mat...

引用

2023 International Conference on Frontiers of Robotics and Software engineering, FRSE 2023

作者： He, Shiming Yi, Tongzhijian Yang, Bo Guo, Chenxi Changsha University of Science & Technology School of Computer and Communication Engineering The Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha China

ISBN: (纸本)9798350301113

Increasingly complex systems contain large numbers of devices that generate great number of multivariate time series that are monitored and recorded. For anomaly detection of these complex time series, deep learning technology is often used to build a learning model for anomaly detection. The method based on deep learning usually takes a long time to train data in practice, and the online service system is deployed or changed very frequently, which makes the online service system need to frequently spend a lot of time updating the anomaly detection model. To overcome this challenge, this paper proposes a fast multivariate time series anomaly detection (FMAD) approach based on matrix completion. After sampling the primitive low-rank matrix, FMAD uses matrix completion to reconstruct the original low-rank matrix to detect anomalies in multivariate time series. We conducted related experimental analysis and discussion. The experimental results on three real data sets show that FMAD has improved the problem of too long model training time and improved the detection efficiency. © 2023 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Adaptive data-Free Quantization

arXiv

引用

arXiv 2023年

作者： Qian, Biao Wang, Yang Hong, Richang Wang, Meng Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer Science and Information Engineering Hefei University of Technology China

data-free quantization (DFQ) recovers the performance of quantized network (Q) without accessing the original data, but generates the fake sample via a generator (G) by learning from full-precision network (P), which, however, is totally independent of Q, overlooking the adaptability of the knowledge from generated samples, i.e., informative or not to the learning process of Q, resulting into the overflow of generalization error. Building on this, several critical questions - how to measure the sample adaptability to Q under varied bit-width scenarios? how to generate the samples with large adaptability to improve Q's generalization? whether the largest adaptability is the best? To answer the above questions, in this paper, we propose an Adaptive data-Free Quantization (AdaDFQ) method, which revisits DFQ from a zero-sum game perspective upon the sample adaptability between two players - a generator and a quantized network. Following this viewpoint, we further define the disagreement and agreement samples to form two boundaries, where the margin is optimized to address the over-and-under fitting issues, so as to generate the samples with adaptive adaptability to Q. Our AdaDFQ reveals: 1) the largest adaptability is NOT the best for sample generation to benefit Q's generalization;2) the knowledge of the generated sample should not be informative to Q only, but also related to the category and distribution information of the training data for P. The theoretical and empirical analysis validate the advantages of AdaDFQ over the state-of-the-arts. Our code is available at https://***/hfutqian/AdaDFQ. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Research on Grape Disease Recognition Based on Improved YOLOv5s in Complex Environment 5

Research on Grape Disease Recognition Based on Improved YOLO...

引用

5th International Conference on Artificial Intelligence Technologies and Applications, ICAITA 2023

作者： Zhou, Zili Gu, Shenming 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

ISBN: (纸本)9781643684840

Aiming at the problems of low recognition accuracy of detector algorithm caused by occlusion interference of grape disease spot in natural complex background, non-significant visual features of grape spot, small target of disease spot and uneven number of sample categories, this study proposed an improved model method based on YOLOv5s. The new WiseIoU is used as the loss function to calculate the positioning loss, so as to improve the accuracy of the algorithm to predict the positioning and speed up the convergence of the model. CA attention mechanism was added to the basic network to enhance grape lesion feature extraction. A small target detection layer was added to the head network of the model to improve the recognition and detection ability of grape small target lesions. IDADP-grape disease recognition image dataset is taken as the main dataset of this study, and combined with self-collected grape disease images, the experimental data set is composed and the training test is carried out. The experimental results show that the improved algorithm mAP0.5 reaches 84.40%, which is 1.30% higher than the baseline model. Compared with other detector algorithms, mAP0.5 and recognition accuracy have improved to different degrees. This method has strong anti-interference robustness and can be used for rapid and accurate identification of grape diseases in complex field environment. © 2024 The Authors.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

Intelligent Innovation dataset on Scientific Research Outcomes and Patents

引用

Journal of Social Computing 2025年第1期6卷 63-73页

作者： Wu, Xinran Zou, Hui Xing, Yidan Qu, Jingjing Guo, Weiwei Fu, Xiaoming School of Computer Science Fudan University Shanghai Key Laboratory of Intelligent Information Processing Shanghai200433 China Shanghai University Department of Library and Information Science Shanghai200444 China Big Data Technology Co. Ltd. Shanghai200082 China Shanghai Artificial Intelligence Laboratory Shanghai200082 China Fudan University Department of Management Shanghai200433 China Fudan University Shanghai Key Laboratory of Intelligent Information Processing Shanghai200433 China Institute of Computer Science University of Göttingen Göttingen37073 Germany

Various stakeholders, such as researchers, government agencies, businesses, and research laboratories require a large volume of reliable scientific research outcomes including research articles and patent data to support their work. These data are crucial for a variety of application, such as advancing scientific research, conducting business evaluations, and undertaking policy analysis. However, collecting such data is often a time-consuming and laborious task. Consequently, many users turn to using openly accessible data for their research. However, these existing open dataset releases typically suffer from lack of relationship between different data sources and a limited temporal coverage. To address this issue, we present a new open dataset, the Intelligent Innovation dataset (IIDS), which comprises six interrelated datasets spanning nearly 120 years, encompassing paper information, paper citation relationships, patent details, patent legal statuses, and funding information. The extensive contextual and extensive temporal coverage of the IIDS dataset will provide researchers and practitioners and policy maker with comprehensive data support, enabling them to conduct indepth scientific research and comprehensive data analyses. © 2020 Tsinghua University Press.

关键词： Decentralized finance

来源：评论

学校读者我要写书评

暂无评论

A Novel 3D Medical Image Segmentation Model Using Improved SAM

A Novel 3D Medical Image Segmentation Model Using Improved S...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Kuang, Yuansen Ma, Xitong Zhao, Jing Wang, Guangchen Zeng, Yijie Liu, Song Key Laboratory of Computing Power Network and Information Security Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China

ISBN: (纸本)9781665410205

3D medical image segmentation is an essential task in the medical image field, which aims to segment organs or tumours into different labels. A number of issues exist with the current 3D medical image segmentation task: existing models cannot simultaneously obtain the space correlation and depth correlation of 3D slices;previous models suffer from local detail loss of positional embedding in 3D images;previous approaches often have blurring of boundaries in segmenting 3D images. To solve these shortcomings, we propose a 3D medical image segmentation model named TPM-SAM. In our model, we design a twinchannel image encoder to simultaneously capture the space correlation and depth correlation of 3D slices through a multi-head attention mechanism and improved adapters. Furthermore, we design a prompt encoding generator, which divides the volumetric image into small blocks and better captures the local detail information. In addition, we introduce a multi-layer aggregation decoder by employing U-Net with multi-level skip connection to solve the blurring of boundaries in processing 3D images. Finally, we experimented and evaluated our model on KiTS21 and LiTS17 datasets to compare with other baseline models. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Rethinking data-Free Quantization as a Zero-Sum Game

arXiv

引用

arXiv 2023年

data-free quantization (DFQ) recovers the performance of quantized network (Q) without accessing the real data, but generates the fake sample via a generator (G) by learning from full-precision network (P) instead. However, such sample generation process is totally independent of Q, specialized as failing to consider the adaptability of the generated samples, i.e., beneficial or adversarial, over the learning process of Q, resulting into non-ignorable performance loss. Building on this, several crucial questions — how to measure and exploit the sample adaptability to Q under varied bit-width scenarios? how to generate the samples with desirable adaptability to benefit the quantized network? — impel us to revisit DFQ. In this paper, we answer the above questions from a game-theory perspective to specialize DFQ as a zero-sum game between two players — a generator and a quantized network, and further propose an Adaptability-aware Sample Generation (AdaSG) method. Technically, AdaSG reformulates DFQ as a dynamic maximization-vs-minimization game process anchored on the sample adaptability. The maximization process aims to generate the sample with desirable adaptability, such sample adaptability is further reduced by the minimization process after calibrating Q for performance recovery. The Balance Gap is defined to guide the stationarity of the game process to maximally benefit Q. The theoretical analysis and empirical studies verify the superiority of AdaSG over the state-of-the-arts. Our code is available at https://***/hfutqian/AdaSG. Copyright © 2023, The Authors. All rights reserved.

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

Legal Risks and Governance Paths for Generative AI - A Case Study of ChatGPT 13

Legal Risks and Governance Paths for Generative AI - A Case ...

引用

13th International Conference on information Technology in Medicine and Education, ITME 2023

作者： Zhenfeng, Wang Quanwang, Lin Huihui, Gao Xiehua, Yu University of Fujian Province Key Laboratory of Big Data and Artificial Intelligence China MinNan Science and Technology University Fujian China School of Computer and Information MinNan Science and Technology University Quanzhou China University of Fujian Key Laboratory of Big Data and Artificial Intelligence Quanzhou China

ISBN: (纸本)9798350319156

The release of OpenAI's ChatGPT chatbot has sparked widespread concern around the world about the threats and legal issues of generative AI. Although this technological innovation has opened a new chapter in the field of artificial intelligence, it has also caused the public to worry about its possible negative effects. In order to solve the above problems and reasonably regulate the legal risks of generative artificial intelligence, this paper mainly adopts the literature, quantitative and qualitative analysis methods, etc. The first part mainly introduces the development process and current situation of generative artificial intelligence, the second part mainly discusses the impact of generative artificial intelligence on tort law, the third part mainly discusses the legal risks of generative artificial intelligence, and the fourth part mainly discusses the governance path of generative artificial intelligence legal risks. The last part is the conclusion of the research. Among them, the fourth part is the focus of this paper. By taking these issues into account, we can provide legal guarantees for the sustainable development of the generative AI industry and ensure that the widespread application of AI benefits human society[1]. © 2023 IEEE.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

SAR Pulse Radio Frequency Interference Suppression Based on Modified Eigen Subspace Projection

SAR Pulse Radio Frequency Interference Suppression Based on ...

引用

2023 International Conference on Remote Sensing, Mapping, and Geographic Systems, RSMG 2023

作者： Liu, Mingyue Shu, Gaofeng Li, Ning School of Computer and Information Engineering Henan University Kaifeng475004 China Henan Key Laboratory of Big Data Analysis and Processing Kaifeng475004 China Henan Province Engineering Research Center of Spatial Information Processing Kaifeng475004 China

ISBN: (纸本)9781510668898

Synthetic aperture radar (SAR), as an active microwave remote sensor with high-resolution earth observation, is easily affected by radio frequency interference (RFI). Pulse radio frequency interference (PRFI), as a typical form of RFI, pollutes SAR raw data with bright stripes, which increases the difficulty of SAR image interpretation. However, the conventional eigen subspace projection (ESP) method decomposes the whole pulse into the interferences and the useful signals, resulting in the loss of useful signals in the non-interference positions with the pulse. To solve this problem, this paper proposes an improved SAR PRFI suppression method based on ESP. Firstly, the receiving pulses containing PRFI are detected and the specific position of PRFI in the pulse are determined by eigenvalue decomposition and energy accumulation in time-frequency domain, respectively. Secondly, the PRFI signal is separated from the useful signal by applying the ESP method only to the data where the PRFI is located. Finally, the PRFI signal projected by ESP is eliminated from the raw data. Simulation results shown that the proposed method can better protected useful signals and improve SAR image quality compared with conventional methods. © 2023 SPIE.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Regional Winter Wheat Yield Estimation Based on CASA Model and Improved Harvest Index Using Remote Sensing Images

Regional Winter Wheat Yield Estimation Based on CASA Model a...

引用

2023 International Conference on Remote Sensing, Mapping, and Geographic Systems, RSMG 2023

作者： Li, Qian Zhao, Jianhui Yang, Huijin Li, Ning School of Computer and Information Engineering Henan University Kaifeng475004 China Henan Province Engineering Research Center of Spatial Information Processing Kaifeng475004 China Henan Key Laboratory of Big Data Analysis and Processing Kaifeng475004 China

ISBN: (纸本)9781510668898

Accurately and timely grasping agricultural information at the regional scale helps to solve food security issues and formulate agricultural policies. Remote sensing images have the advantages of wide monitoring range and the ability to eliminate human interference. In recent years, they have been increasingly valued in crop yield estimation. The CASA model is used to estimate the net primary productivity (NPP) of crops and then combine it with the harvest index (HI) to estimate crop yields. However, most studies use a fixed HI for yield estimation, which can lead to low accuracy. In this context, the HI is studied and improved for winter wheat yield prediction in this research. The main contents are as follows: Firstly, time-series Sentinel-2 optical data and meteorological data were used as inputs of the CASA model to calculate the NPP of winter wheat. Then, HI was calculated using the fieldwork data and a method was proposed to improve it due to its geographic location-related characteristics. Finally, the calculated NPP and improved HI were used to estimate the winter wheat yield. The experimental results prove that the proposed method has higher accuracy than original methods, with determination coefficient(R2) of 0.595, root mean square error(RMSE) of 793.4 kg/ha, and mean absolute error(MAE) of 659.53 kg/ha. © 2023 SPIE.

关键词： Crops

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：