检索结果-内蒙古大学图书馆

Intelligent signal processing and Effective Communication Technologies (INSPECT), IEEE International Conference on

作者： Mithlesh Arya Arvind Singh Rajpoot Dhruv Pathak Bhagyansh Garg Megha Gupta Abha Jain Computer Science & Engineering Swami Keshvanand Institute of Technology Management & Gramothan Jaipur India Computer Science & Engineering Sardar Vallabhbhai National Institute of Technology Surat India Computer Science & Engineering (DS) Swami Keshvanand Institute of Technology Management & Gramothan Jaipur India

ISBN: (数字)9798350379525

ISBN: (纸本)9798350379532

Lung cancer is most common in men in India. Smoking is the main cause of lung cancer, apart from this there are many other reasons also. There are many tests to detect lung cancer such as chest X-ray and Computed tomography (CT). In this paper we are using two datasets to validate our work. First dataset-I is numeric data which is collectively based on surveys and second dataset-ii contains CT images of lungs. In numeric data we have 15 features and have applied a neural network and ensemble classifier to classify the data into normal and cancerous classes. After that Pearson correlation method is used to select the most prominent features. The highest performance is obtained when the XGBoost ensemble classifier is used and the obtained values are Precision (95%), Recall (98%) and Fl score (96%). In the second dataset-ii we have applied CNN to classify the lung images into multiple classes (normal, benign and malignant). But the accuracy is not prominent. Then we are using feature selection methods on flatten images and applying ensemble classifiers (Random forest, Gradient Boosting and Support Vector Classifier) to classify the images multiple classes. Prediction after feature selection is given a more prominent result and these are precision(1.00), recall(1.00) and F1 score(1.00).

关键词： Surveys Accuracy Computed tomography Lungs neural networks Lung cancer Feature extraction Ensemble learning X-ray imaging image classification

来源：评论

学校读者我要写书评

暂无评论

Cuffless blood pressure estimation from photoplethysmography using deep convolutional neural network and transfer learning

引用

BIOMEDICAL signal processing AND CONTROL 2024年 93卷

作者： Koparir, Hueseyin Murat Arslan, Ozkan Tekirdag Namik Kemal Univ Dept Elect & Commun Engn TR-59860 Tekirdag Turkiye

Blood pressure (BP) monitoring is an essential indicator for diseases of the cardiovascular system. Early detection of abnormalities in BP helps to significantly reduce the risk of diseases such as chronic heart failure, stroke and hypertension. In this study, we propose a new framework for systolic, diastolic and mean arterial blood pressure (SBP, DBP and MBP) estimation using PPG signals and its derivatives. The proposed framework includes the extraction of deep features by pre-trained CNN models based on transfer learning using images of signal waveforms. The most distinctive features are obtained by using the RFE feature selection algorithm on the extracted deep features. The selected deep features are exposed to a range of well-known machine and deep learning regression methods. The proposed BP estimation models have been evaluated with statistical metrics, visual analytical tools and international gold standards such as AAMI and BHS. Experimental results show that the first derivative of PPG, the VPG input image, the deep features obtained with DenseNet121 and the bidirectional gated recurrent unit (Bi-GRU) algorithm provide the best BP estimation performance. The results reveal that the SBP, DBP and MBP estimation models are grade-A according to the BHS protocol and meet the AAMI standard. The presented framework has been compared with well-established studies using the MIMIC ii dataset, and the comparative analysis confirms that the proposed method outperforms existing techniques. The deep learning model for BP estimation offers a highly accurate, fast and practical system, especially in the monitoring of patients at risk of hypertension.

关键词： Photoplethysmography Blood pressure Cuffless and non -invasive measurement Transfer learning Machine learning Deep learning

来源：评论

学校读者我要写书评

暂无评论

Learning the image Prior by Unrolling an Optimization Method

Learning the Image Prior by Unrolling an Optimization Method

引用

European signal processing Conference (EUSIPCO)

作者： Silvia Bonettini Giorgia Franchini Danilo Pezzi Marco Prato Department of Informatics Physics and Mathematics University of Modena and Reggio Emilia Modena Italy

ISBN: (数字)9789082797091

ISBN: (纸本)9781665467995

Nowadays neural networks are omnipresent thanks to the amazing adaptability they possess, despite their poor interpretability and the difficulties they give when manipulating the parameters. On the other side, we have the classical variational approach, where the restoration is obtained as the solution of a given optimization problem. The bilevel approach is connected to both approaches and consists first in devising a parametric formulation of the variational problem, then in optimizing these parameters with respect to a given dataset of training data. In this work we analyze the classical bilevel approach in combination with unrolling techniques, where the parameters of the variational problem are trained with respect to the results obtained after a fixed number of iterations of an optimization method applied to it. This results in a large scale optimization problem which can be solved by means of stochastic methods; as we observed in our numerical experiments, the stochastic approach can produce medium accuracy results in very few epochs. Moreover, our experiments also show that the unrolling approach leads to results which are comparable with those of the original bilevel method in terms of accuracy.

关键词： neural networks Optimization methods stochastic processes Training data Europe signal processing Minimization

来源：评论

学校读者我要写书评

暂无评论

Adadb: Adaptive Diff-Batch Optimization Technique for Gradient Descent

引用

IEEE ACCESS 2021年 9卷 99581-99588页

作者： Khan, Muhammad U. S. Jawad, Muhammad Khan, Samee U. COMSATS Univ Islamabad Dept Comp Sci Abbottabad Campus Abbottabad 22060 Pakistan COMSATS Univ Islamabad Dept Elect Engn Lahore Campus Lahore 54000 Pakistan Mississippi State Univ Elect & Comp Engn Starkville MS 39762 USA

Gradient descent is the workhorse of deep neural networks. Gradient descent has the disadvantage of slow convergence. The famous way to overcome slow convergence is to use momentum. Momentum effectively increases the learning factor of gradient descent. Recently, many approaches have been proposed to control the momentum for better optimization towards global minima, such as Adam, diffGrad, and AdaBelief. Adam decreases the momentum by dividing it with square root of moving averages of squared past gradients or second moment. The sudden decrease in the second moment often results in the overshoot of the gradient from the minima and then settle at the closest minima. DiffGrad decreases this problem by using a friction constant based on the difference of current gradient and immediate past gradient in Adam. The friction constant further decreases the momentum and results in slow convergence. AdaBelief adapts the step size according to the belief in the current gradient direction. Another famous way of fast convergence is to increase the batch size adaptively. This paper proposes a new optimization technique named adaptive diff-batch or adadb that removes the problem of overshooting gradient in Adam, slow convergence in diffGrad, and combines the methods with adaptive batch size for further increase in convergence rate. The proposed technique uses the friction constant based on the past three differences of gradients rather than one as in diffGrad and a condition to decide the use of friction constant. The proposed technique has outperformed the Adam, diffGrad, and AdaBelief optimizers on synthetic complex non-convex functions and real-world datasets.

关键词： Convergence Friction Optimization neural networks Training stochastic processes signal processing algorithms Machine learning gradient descent optimization image classification

来源：评论

学校读者我要写书评

暂无评论

Multi-task multi-scale attention learning-based facial age estimation

引用

IET signal processing 2023年第2期17卷

作者： Shi, Chaojun Zhao, Shiwei Zhang, Ke Feng, Xiaohan North China Elect Power Univ Dept Elect & Commun Engn Baoding Hebei Peoples R China North China Elect Power Univ Hebei Key Lab Power Internet Things Technol Baoding Hebei Peoples R China

Face-based age estimation strongly depends on deep residual networks (ResNets), used as the backbone in the relevant research. However, ResNet-based methods ignore the importance of some large-scale facial information and other facial age attributes. Inspired by the attention mechanism, a multi-task learning framework for face-based age estimation called multi-task multi-scale attention is proposed. First, the authors embed the alternative strategy structure of dilated convolution into ResNet34 to construct a multi-scale attention module (MSA) to improve the network's receptive field, which extracts local age-sensitive information while obtaining multi-scale features. The MSA can have a larger receptive field to extract both large-scale and local detailed feature information. Second, multi-task learning network structures are built to predict gender and race, which can share rigid network parameters to improve age estimation and improve the accuracy of age estimation by other age-related parameters. Finally, the Kullback-Leibler divergence loss is adopted between a Dirac delta label and a Gaussian prediction to guide the training. The numerical tests on the MORPH Album ii and Adience datasets prove the superiority of the proposed method over other state-of-the-art ones.

关键词： computer vision feature selection feedforward neural nets image processing neural nets

来源：评论

学校读者我要写书评

暂无评论

REGULARIZING neural NETWORKS BY stochasticALLY TRAINING LAYER ENSEMBLES 30

REGULARIZING NEURAL NETWORKS BY STOCHASTICALLY TRAINING LAYE...

引用

30th IEEE International Workshop on Machine Learning for signal processing (MLSP)

作者： Labach, Alex Valaee, Shahrokh Univ Toronto Dept Elect & Comp Engn Toronto ON Canada

ISBN: (纸本)9781728166629

Dropout and similar stochastic neural network regularization methods are often interpreted as implicitly averaging over a large ensemble of models. We propose STE (stochastically trained ensemble) layers, which enhance the averaging properties of such methods by training an ensemble of weight matrices with stochastic regularization while explicitly averaging outputs. This provides stronger regularization with no additional computational cost at test time. We show consistent improvement on various image classification tasks using standard network topologies.

关键词： neural networks regularization dropout model averaging ensemble methods

来源：评论

学校读者我要写书评

暂无评论

An efficient deep learning framework for P300 evoked related potential detection in EEG signal

引用

COMPUTER methods AND PROGRAMS IN BIOMEDICINE 2023年 229卷 107324-107324页

作者： Havaei, Pedram Zekri, Maryam Mahmoudzadeh, Elham Rabbani, Hossein Isfahan Univ Technol Dept Elect & Comp Engn Esfahan *** Iran Isfahan Univ Med Sci Med Image & Signal Proc Res Ctr Esfahan Iran Isfahan Univ Med Sci Sch Adv Technol Med Esfahan Iran Isfahan Univ Technol Dept Elect & Comp Engn Esfahan Iran

Background: Incorporating the time-frequency localization properties of Gabor transform (GT), the com-plexity understandings of convolutional neural network (CNN), and histogram of oriented gradients (HOG) efficacy in distinguishing positive peaks can exhibit their characteristics to reveal an effective solution in the detection of P300 evoked related potential (ERP). By applying a drastic number of convolutional lay-ers, the majority of deep networks elicit sufficient properties for the output determination, leading to gigantic and time-consuming structures. In this paper, we propose a novel deep learning framework by the combination of tuned GT, and modified HOG with the CNN as "TGT-MHOG-CNN" for detection of P300 ERP in EEG ***: In the proposed method, GT is tuned based on triangular function for EEG signals, and then spectrograms including time-frequency information are captured. The function's parameters are justified to differentiate the signals with the P300 component. Furthermore, HOG is modified (MHOG) for the 2-D EEG signal, and consequently, gradients patterns are extracted for the target potentials. MHOG is potent in distinguishing the positive peak in the general waveform;however, GT unravels time-frequency infor-mation, which is ignored in the gradient histogram. These outputs of GT and MHOG do not share the same nature in the images nor overlap. Therefore, more extensive information is reached without re-dundancy or excessive information by fusing them. Combining GT and MHOG provides different patterns which benefit CNN for more precise detection. Consequently, TGT-MHOG-CNN ends in a more straightfor-ward structure than other networks, and therefore, the whole performance is acceptable with faster rates and very high ***: BCI Competition ii and iiI datasets are used to evaluate the performance of the proposed method. These datasets include a complete record for P300 ERP with BCI20 0 0 using a paradigm, and it has nu-merous noise

关键词： Convolutional neural network Gabor transform Modified histogram of oriented gradients P300 detection

来源：评论

学校读者我要写书评

暂无评论

Implicit Bias Analysis in The Training of Compact neural Networks For Inverse Problems

Implicit Bias Analysis in The Training of Compact Neural Net...

引用

Emerging Trends in Industry 4.0 (ETI 4.0)

作者： Uma Reddy S Vinod Kumar Aakriti Yadav Amandeep Nagpal Ashwani Kumar Nashat Ali Soud Department of Artificial Intelligence and Machine Learning New Horizon College of Engineering Bangalore Institute of Aeronautical Engineering Dundigal Hyderabad DoCSE GyanGanga Institute of Technology and Sciences Jabalpur M.P India Lovely Professional University Phagwara Lloyd Institute of Engineering & Technology Knowledge Park II Greater Noida Uttar Pradesh Collage of Pharmacy National University of Science and Technology Dhi Qar Iraq

ISBN: (数字)9798350373783

ISBN: (纸本)9798350373790

This paper suggests three new techniques to improve tiny neural networks’ inverse problem-solving. Traditional approaches, like basic linear regression models, lack the flexibility and bias reduction needed to handle complex real-world challenges. Our recommended methods—BRT, ABN, and LFD—can solve these issues. The BRT technique balances bias reduction with data fitting by using L1 and L2 regularization to handle hidden biases. Change batch normalization parameters on the fly with ABN for additional possibilities. This optimizes feature scaling and accuracy while reducing false positives. But LFD increases F1 scores by applying several loss functions. This is important for avoiding false positives and negatives. Some of our experiments used phony data, yet they were positive. The comparative analysis reveals that recommended solutions outperform regular techniques on key performance parameters. ABN is precise, LFD is accurate, and BRT has an average F1 number. These results demonstrate the value of new ideas tailored to each assignment. However, our study supports the real-world use of these procedures or their modifications. By improving tiny neural networks’ performance, medical diagnostics, signal reconstruction, and picture processing might become more accurate and precise.

关键词： Training Technological innovation Accuracy image processing neural networks Transforms signal reconstruction Problem-solving Batch normalization Monitoring

来源：评论

学校读者我要写书评

暂无评论

Exploring the use of Optical Domain Information for Automated Target Detection and Recognition 2

Exploring the use of Optical Domain Information for Automate...

引用

Conference on Virtual, Augmented, and Mixed Reality (XR) Technology for Multi-Domain Operations ii

作者： Hickman, Duncan L. Tektonex Ltd Argyll Rd Kilcreggan Argyll & Bute Scotland

ISBN: (数字)9781510643567

ISBN: (纸本)9781510643567

Automated Target Detection and Recognition (ATD/R) of targets in a cluttered background presents major challenges for the designers of military sensor systems. To achieve acceptable levels of detection and recognition performance, extensive use is often made of complex image processing techniques such as neural network architectures which attempt to replicate the capabilities of the human vision system. Although such methods generate good levels of performance, they are often not suitable for those applications where smaller and lower cost sensor systems are used across a more diverse range of scene content and imaging conditions. A new approach is proposed here where system performance is achieved through a more effective balance between optical domain and post-detection processing. Specifically, the signal to Clutter Ratio (SCR) is maximised by using broadband spectral and polarisation information to offset performance deficiencies associated with simple image and data processing functions. It is shown that this approach offers a basis for introducing ATD/R functionality in low cost imaging systems such as those flown on drones. A simple imaging system is used to demonstrate the concept, which compromises two broadband cameras operating in the visible and near infrared bands, and with one of the sensors additionally providing polarimetric information. The concept of a joint spectral polarisation weight map is proposed and the potential performance gain is illustrated using targets in moderate to high clutter situations. The results obtained indicate potential benefits for future ATD/R systems and it is hoped that this will encourage future design engineers to consider the wider use of optical domain information.

关键词： Target detection and recognition ATD/R spectral processing polarimetry colour weight map polarisation weight map joint weight map signal to clutter optical domain processing

来源：评论

学校读者我要写书评

暂无评论

Use of images Augmentation and Implementation of Doubly stochastic Models for Improving Accuracy of Recognition Algorithms Based on Convolutional neural Networks

Use of Images Augmentation and Implementation of Doubly Stoc...

引用

Conference on Systems of signal Synchronization, Generating and processing in Telecommunications (SYNCHROINFO)

作者： Dementyiev, V. E. Andriyanov, N. A. Vasilyiev, K. K. Ulyanovsk State Tech Univ Ulyanovsk Russia JSC RPC Istok Fryazino Russia

ISBN: (纸本)9781728160726

The article proposes methods for increasing the efficiency of pattern recognition in images using convolutional neural networks in conditions of insufficient reference images. It is proposed to use various kinds of image transformations to generate new images and increase the training sample volume. In particular, methods of scaling, adding noise, blurring, etc. are proposed. In addition to standard image transformations, it is supposed to increase the database by presenting patterns using a doubly stochastic model. In addition, the need for the use of regularization in training is noted. The recognition results are compared for two convolutional neural networks that differ only in the number of neurons for different initial data. It is shown that the highest recognition accuracy is provided by a network trained on a large number of augmentations in combination with images generated by a doubly stochastic model.

关键词： convlolutional neural networks doubly stochastic model recognition accuracy data augmentation drop out regularization

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：