检索结果-内蒙古大学图书馆

Application of convolution neural network for Adaptive Traffic Controller System

KSCE JOURNAL OF CIVIL ENGINEERING 2022年第9期26卷 4062-4072页

作者： Ahmed, Muaid Abdulkareem Alnazir Khoo, Hooi Ling Ng, Oon-Ee Univ Tunku Abdul Rahman Lee Kong Chian Fac Engn & Sci Dept Civil Engn Bandar Sungai Long 43000 Kajang Malaysia Univ Tunku Abdul Rahman Lee Kong Chian Fac Engn & Sci Dept Mechatron & Biomed Engn Kajang 43000 Malaysia

This paper presents an adaptive traffic control system based on deep convolution neural network (DCNN) technique for a multimodal traffic environment. The deep learning controller technique is based on feature mapping to estimate the optimal state-action value function. The controller executes acyclic phase assignment based on minimum green time duration for traffic signal operation. Moreover, the research implements Discrete Lane Cells (DLC) approach for state representation. A number of traffic performance measures were examined and selected traffic conditions were tested. The findings indicate that the DCNN system has superior performance in over-saturated traffic condition. The DCNN agent has achieved significant 85% to 95% lower waiting time, l7% to 38% shorter travel time, and it has mitigated the highest median flow rate at 295 veh/s in comparison to other traffic systems. For under-saturated test scenarios, a fair comparable performance is measured for the proposed controller. Overall, the DCNN system provides stable performance across all the tested signal junction scenarios in comparison to other controller systems.

关键词： Adaptive traffic control system convolution neural network Deep reinforcement learning Multimodal environment Intersection management Traffic signal operation

来源：评论

学校读者我要写书评

暂无评论

An Emotion Analysis Method Using Multi-Channel convolution neural network in Social networks

引用

Computer Modeling in Engineering & Sciences 2020年第10期125卷 281-297页

作者： Xinxin Lu Hong Zhang School of Computer Science and Technology Zhoukou Normal UniversityZhoukou466001China School of Network Engineering Zhoukou Normal UniversityZhoukou466001China

As an interdisciplinary comprehensive subject involving multidisciplinary knowledge,emotional analysis has become a hot topic in psychology,health medicine and computer *** has a high comprehensive and practical application *** research based on the social network is a relatively new topic in the field of psychology and medical health *** text emotion analysis of college students also has an important research significance for the emotional state of students at a certain time or a certain period,so as to understand their normal state,abnormal state and the reason of state change from the information they *** view of the fact that convolutional neural network cannot make full use of the unique emotional information in sentences,and the need to label a large number of highquality training sets for emotional analysis to improve the accuracy of the model,an emotional analysismodel using the emotional dictionary andmultichannel convolutional neural network is proposed in this ***,the input matrix of emotion dictionary is constructed according to the emotion information,and the different feature information of sentences is combined to form different network input channels,so that the model can learn the emotion information of input sentences from various feature representations in the training ***,the loss function is reconstructed to realize the semi supervised learning of the ***,experiments are carried on COAE 2014 and self-built data *** proposed model can not only extract more semantic information in emotional text,but also learn the hidden emotional information in emotional *** experimental results show that the proposed emotion analysis model can achieve a better classification *** with the best benchmark model gram-CNN,the F1 value can be increased by 0.026 in the self-built data set,and it can be increased by 0.032 in the COAE 2014 data set.

关键词： Emotion analysis model emotion dictionary convolution neural network semi supervised learning deep learning pooling feature feature mapping

来源：评论

学校读者我要写书评

暂无评论

Evaluation method for the corona discharge of insulator based on convolution neural network with the dual-spectra camera

引用

OPTICAL ENGINEERING 2020年第10期59卷

作者： Wang, Yan Qian, Yunsheng Nanjing Univ Sci & Technol Sch Elect Engn & Photoelect Technol Nanjing Peoples R China

As insulators play an important role in power transmission lines, we propose an intelligent method based on the convolution neural network (CNN) to evaluate the corona discharge of an insulator. In this method, the imaging device is a dual-spectra camera with a visible channel and an ultraviolet (UV) channel. The CNN is adopted to identify the detection distance of the insulator with the visible channel. To train the network, the dataset of the insulator is obtained by the experimental setup and deep convolutional generative adversarial networks. Through adjusting the training parameters and optimizing the network structure, an optimal trained model is achieved. Then the image pixel ratio method is adopted to measure the UV signal strength of the images captured by the UV channel. Meanwhile, the relationship between the detection distance and the UV signal strength is discussed. The critical value for the corona discharge of the insulator is obtained via experiments at the standard detection distance. Finally, the corona discharge of the insulator is evaluated by combining the detection distance with the UV signal strength. The experimental results show the method has the advantages of high accuracy and robustness and can effectively evaluate the corona discharge of the insulator. (C) 2020 Society of Photo-Optical Instrumentation Engineers (SPIE)

关键词： convolution neural network corona discharge dual-spectra camera

来源：评论

学校读者我要写书评

暂无评论

Quantum deep learning in Parkinson's disease prediction using hybrid quantum-classical convolution neural network

引用

QUANTUM INFORMATION PROCESSING 2024年第12期23卷 1-32页

作者： Sha, Mohemmed Rahamathulla, Mohamudha Parveen Prince Sattam Bin Abdulaziz Univ Coll Comp Engn & Sci Dept Software Engn Al Kharj Saudi Arabia Prince Sattam Bin Abdulaziz Univ Coll Med Dept Basic Med Sci Al Kharj Saudi Arabia

Deep learning, also known as DL, holds great potential within the field of artificial intelligence. Fast problem-solving approaches are widely used in quantum computing. Large multidimensional space is utilized to categorize and address intricate problems. The different algorithms have the ability to interact in a space with multiple dimensions and find solutions to the problems. Quantum deep learning facilitates different mining procedures by incorporating precise advancements in quantum computing. Prompt and accurate identification during the early stages of progression is crucial for various severe and life-threatening illnesses like cancer, hepatotoxicity, cardio toxicity, nephrotoxicity, and others. Currently, there is a critical need to create rapid, precise, and highly effective approaches for predicting different diseases. These methods should also be feasible and nonintrusive. Dementia, a highly hazardous condition, has a significant impact on the human nervous system. Dementia often includes Parkinson's as one of its prominent symptoms. The patient's entire operational behavior will be impacted. The proposed system is utilizing machine learning and quantum computing to develop a method for predicting Parkinson's disease based on speech signals. Quantum computers can be used to assist in identifying cancer by using a hybrid quantum-classical convolution neural network (QCCNN). This network is inspired by convolution neural networks (CNNs) but has been modified for quantum computing in order to improve the process of mapping features. Dimensionality reduction algorithms, principal component analysis (PCA) are applied to the preprocessed dataset to make predictions about diseases. The standard dataset from UCI machine learning repository will be used to determine the performance of the model. Ensemble models exceed the precision of highly accurate techniques such as neural networks. To demonstrate the superior detection capability of our model, we have compar

关键词： Parkinson's disease Quantum computing Dimensionality reduction convolution neural network Deep learning

来源：评论

学校读者我要写书评

暂无评论

Deep convolution neural network for Big Data Medical Image Classification

引用

IEEE ACCESS 2020年 8卷 105659-105670页

作者： Ashraf, Rehan Habib, Muhammad Asif Akram, Muhammad Latif, Muhammad Ahsan Malik, Muhammad Sheraz Arshad Awais, Muhammad Dar, Saadat Hanif Mahmood, Toqeer Yasir, Muhammad Abbas, Zahoor Natl Text Univ Dept Comp Sci Faisalabad 37610 Pakistan Balochistan Univ Informat Technol Engn & Manageme Dept Software Engn Quetta 87300 Pakistan Univ Agr Faisalabad Dept Comp Sci Faisalabad 38000 Pakistan Govt Coll Univ Faisalabad Dept Informat Technol Faisalabad 38000 Pakistan Govt Coll Univ Faisalabad Dept Software Engn Faisalabad 38000 Pakistan Mirpur Univ Sci & Technol MUST Dept Software Engn Mirpur 10250 Pakistan Univ Engn & Technol Lahore Dept Comp Sci Faisalabad 38090 Pakistan

Deep learning is one of the most unexpected machine learning techniques which is being used in many applications like image classification, image analysis, clinical archives and object recognition. With an extensive utilization of digital images as information in the hospitals, the archives of medical images are growing exponentially. Digital images play a vigorous role in predicting the patient disease intensity and there are vast applications of medical images in diagnosis and investigation. Due to recent developments in imaging technology, classifying medical images in an automatic way is an open research problem for researchers of computer vision. For classifying the medical images according to their relevant classes a most suitable classifier is most important. Image classification is beneficial to predict the appropriate class or category of unknown images. The less discriminating ability and domain-specific categorization are the main drawbacks of low-level features. A semantic gap that exists between features of low-level as machine understanding and features of human understanding as high-level perception. In this research, a novel image representation method is proposed where the algorithm is trained for classifying medical images by deep learning technique. A pre-trained deep convolution neural network method with the fine-tuned approach is applied to the last three layers of deep neural network. The results of the experiment exhibit that our method is best suited to classify various medical images for various body organs. In this manner, data can sum up to other medical classification applications which supports radiologist's efforts for improving diagnosis.

关键词： Medical image classification pre-trained DCNN convolution neural network big data image analysis image enhancement biomedical image processing deep learning

来源：评论

学校读者我要写书评

暂无评论

Automatic Sleep Staging Based on Contextual Scalograms and Attention convolution neural network Using Single-Channel EEG

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2024年第2期28卷 801-811页

作者： Wei, Yu Zhu, Yongpeng Zhou, Yihan Yu, Xiaokang Luo, Yuxi Sun Yat Sen Univ Sch Biomed Engn Shenzhen Campus Shenzhen 518100 Guangdong Peoples R China Sun Yat Sen Univ Key Lab Sensing Technol & Biomed Instruments Guang Shenzhen 518100 Guangdong Peoples R China

Single-channel EEG based sleep staging is of interest to researchers due to its broad application prospect in daily sleep monitoring recently. We proposed using contextual scalograms as input and developed a convolutional neural network with attention modules named Co-ScaleNet for sleep staging. The contextual scalograms were obtained by combining the same color channels of three original RGB scalograms from consecutive epochs, and a simple and efficient data augmentation was designed according to their various forms. The Co-ScaleNet consists of two main parts. Firstly, three parallel convolutional branches with attention modules correspondingly extract and fuse features from contextual scalograms at the top layers. The remaining part is a stack of lightweight blocks. We achieved an overall accuracy of 87.0% for healthy individuals, 84.7% for depressed patients. And we obtained comparable performance on the public Sleep-EDFx (82.8%), ISRUC (84.6%) and SHHS datasets (87.7%), including a high recall of N1. The contextual scalograms of R channel as input achieved the best performance, which conform to the features of interest in visual scoring. The attention modules improved the recall of N1 and N3. Overall, the contextual scalograms provided a novel scheme for both contextual information extraction and data augmentation. Our study successfully expanded its application to depression datasets, as well as patients with sleep apnea, demonstrating its wide applicability.

关键词： Sleep Electroencephalography Depression Recording Brain modeling Continuous wavelet transforms Manuals Automatic sleep staging single channel EEG contextual scalograms convolution neural network

来源：评论

学校读者我要写书评

暂无评论

Identify videos with facial manipulations based on convolution neural network and dynamic texture

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2022年第30期81卷 43441-43466页

作者： Megahed, Amr Han, Qi Harbin Inst Technol Sch Comp Sci & Technol Harbin 150080 Peoples R China

Recent facial manipulation techniques based on deep learning can create a highly realistic face by changing expression, attributes, identity, or creating an entire face synthesis, that called recently Deep-Fake. With the rapid appearance of such applications, they have raised great security concerns. Therefore, corresponding forensic techniques are proposed to tackle this issue. However, existing techniques are either based on complex deep networks with a binary classification that are unable to distinguish between facial manipulation types or rely on fragile hand-crafted features with unsatisfactory results. To overcome these issues, we propose a learning-based detection method by creating an uncomplicated CNN network called FMD-Net relying on the dynamic textures as input. Moreover, it is able to distinguish between facial manipulation types such as Deepfake, Face2Face, FaceSwap, and neuralTexture. By using dynamic textures of each video shot, motion and appearance features are combined which helped the network learn manipulation artifacts and provides a robust performance at various compression levels. We conduct extensive experiments on various benchmark datasets (FaceForensics++, DFDC, and Celeb-DF) to empirically demonstrate the superiority and effectiveness of the proposed method with both binary classification and multi-classification against the state-of-the-art methods.

关键词： Face forensics Video manipulation detection convolution neural network Facial manipulation detection Spatiotemporal features

来源：评论

学校读者我要写书评

暂无评论

Early quadtree with nested multitype tree partitioning algorithm based on convolution neural network for the versatile video coding standard

引用

JOURNAL OF ELECTRONIC IMAGING 2024年第4期33卷

作者： Abdallah, Bouthaina Ben Jdidia, Sonda Belghith, Fatma Ben Ayed, Mohamed Ali Masmoudi, Nouri Univ Sfax Natl Engn Sch Sfax Elect & Informat Technol Lab Sfax Tunisia Univ Sfax Sfax Natl Sch Elect & Commun New Technol & Telecom Syst Lab Sfax Tunisia

The Joint Video Experts Team has recently finalized the versatile video coding (VVC) standard, which incorporates various advanced encoding tools. These tools ensure great enhancements in the coding efficiency, leading to a bitrate reduction up to 50% when compared to the previous standard, high-efficiency video coding. However, this enhancement comes at the expense of high computational complexity. Within this context, we address the new quadtree (QT) with nested multitype tree partition block in VVC for all-intra configuration. In fact, we propose a fast intra-coding unit (CU) partition algorithm using various convolution neural network (CNN) classifiers to directly predict the partition mode, skip unnecessary split modes, and early exit the partitioning process. The proposed approach first predicts the QT depth at a CU of size 64 x 64 by the corresponding CNN classifier. Then four CNN classifiers are applied to predict the partition decision tree at a CU of size 32 x 32 using multithreshold values and ignore the rate-distortion optimization process to speed up the partition coding time. Thus the developed method is implemented on the reference software VTM 16.2 and tested for different video sequences. The experimental results confirm that the proposed solution achieves an encoding time reduction of about 46% in average, reaching up to 67.3% with an acceptable increase in bitrate and an unsignificant decrease in quality. (c) 2024 SPIE and IS&T

关键词： versatile video coding computational complexity quadtree with nested multitype tree convolution neural network encoding time reduction

来源：评论

学校读者我要写书评

暂无评论

Atmospheric Turbulence Strength Estimation Using convolution neural network

IEEE PHOTONICS JOURNAL

引用

IEEE PHOTONICS JOURNAL 2023年第6期15卷 1页

作者： Gao, Siyu Liu, Xiaoyun Chen, Yonghao Jiang, Jinyang Liu, Ying Jiang, Yueqiu Shenyang Ligong Univ Sch Sci Dept Genet Engn Shenyang 110159 Peoples R China Shenyang Ligong Univ Dept Dev & Planning Shenyang 110159 Peoples R China

Laser beam transmission in atmospheric turbulence causes image distortion and affects the quality of information transmission in the field of optical communication. The strength of the atmospheric turbulence, which can be characterized by refractive index structure constant C-n(2), significantly influences the properties of a laser beam. The accurate estimation of C-n(2) is essential for understanding the strength of turbulence. Although multilayer perceptron (MLP) and deep neural network (DNN) has been applied to estimate the atmospheric turbulence strength, the estimation accuracy is sensitive to the strength of the turbulence. In this article, we propose a method based on the convolution neural network (CNN) approach to estimate C-n(2) ranging from 10(-17) to 10(-13 )m(-2/3). We experimentally demonstrate that the correlation coefficient (R-2) of the model is 99.39%. The mean relative error (MRE), root mean square error (RMSE), and mean absolute error (MAE) are 0.0047, 0.0916, and 0.0684, respectively. For the turbulence strength with the same order of refractive index structure constant C-n(2), the estimation accuracy of the weak turbulence is higher than that of medium and strong turbulence. Moreover, the mix training different levels of turbulence strength improves the estimation accuracy of C-n(2) compared to that with the same order of C-n(2). Based on the high estimation accuracy of the CNN in the scheme, the proposed method will be able to provide a way of estimating the strength of atmospheric turbulence without the need for additional optical devices.

关键词： Atmospheric turbulence convolution neural network refractive index structure constant C-n(2)

来源：评论

学校读者我要写书评

暂无评论

Multimodal emotion recognition based on manifold learning and convolution neural network

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2022年第23期81卷 33253-33268页

作者： Zhang, Yong Cheng, Cheng Zhang, YiDie Huzhou Univ Sch Informat Engn Huzhou 313000 Peoples R China Liaoning Normal Univ Sch Comp & Informat Technol Dalian 116081 Peoples R China

Multimodal emotion recognition task based on physiological signals is becoming a research hotspot. Traditional methods need to design and extract a series of features from single-channel or multi-channel physiological signals on the basis of extensive domain knowledge. These methods cannot make full use of the relevant information among channels, and the emotion recognition of a single modality cannot fully express the emotional state. This paper proposes a multimodal emotion recognition model based on manifold learning and a convolutional neural network (CNN). The electroencephalograph (EEG) signals are combined with peripheral physiological signals and eye movement signals respectively, the multivariate synchrosqueezing transform (MSST) is used to simulate the joint oscillation structure of multi-channel signals, and then the related feature parameters are extracted and fused into feature vectors. The proposed method finds the corresponding low dimensional embedding features for given high-dimensional features by an improved manifold learning method, which feeds into the deep convolutional neural network (DCNN) model for emotion recognition. We perform extensive four-category experiments on the dimensions of arousal and valence. Results indicate that our proposed model achieves average accuracies of 90.05% and 88.17% on the DEAP and MAHNOB-HCI datasets respectively, which both receive better performances than most of the compared studies, verifying the effectiveness of the model.

关键词： Multimodal Emotion recognition Physiological signals Manifold learning convolution neural network

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：