检索结果-内蒙古大学图书馆

Dynamic Effects of a MCNN-CS Under Electromagnetic Induction and Its Application

INTERNATIONAL JOURNAL OF BIFURCATION AND CHAOS 2022年第11期32卷 2250165-2250165页

作者： Zhang, Fengqing Shi, Hang Ji'e, Musha Wang, Lidan Duan, Shukai Southwest Univ Sch Elect & Informat Engn Chongqing 400715 Peoples R China Southwest Univ Coll Artificial Intelligence Chongqing 400715 Peoples R China

Cellular neural network (CNN) is one of the best artificial neural network models due to its hardware implementability and many applications for image processing. The memristor-based cellular neural network (MCNN) combines the distinct benefits of memristors and neural networks, and it excels in mimicking chaotic systems under the effect of electromagnetic induction. In this work, a novel CNN chaotic system based on a flux-controlled memristor (MCNN-CS) with electromagnetic induction effects is constructed. Dynamical behaviors are examined by controlling the magnetic flux leakage strength and the intensity of the electromagnetic induction related to the memristor. Based on stability analyses and numerical simulation, forward and reverse period-doubling bifurcations, various coexisting attractors, multiperiods, chaos, several periodic windows, and chaotic crisis are observed. Besides, a number of complicated phenomena, including transient chaos, intermittent chaos, sustained chaos and bistability are also observed, demonstrating that this chaotic system has a wide range of dynamic properties. Finally, we generate pseudo-random number sequences based on this new system and apply them to image encryption. The obtained results show that this chaotic system offers superior randomness and is hence appropriate for image encryption applications.

关键词： Memristor Cellular neural Network (CNN) electromagnetic induction effect dynamic analysis image encryption

来源：评论

学校读者我要写书评

暂无评论

Efficient Deep Model-Based Optoacoustic image Reconstruction

Efficient Deep Model-Based Optoacoustic Image Reconstruction

引用

IEEE International Symposium on applications of Ferroelectrics (ISAF)

作者： Christoph Dehner Guillaume Zahnd iThera Medical GmbH Munich Germany

ISBN: (数字)9798350371901

ISBN: (纸本)9798350371918

Clinical adoption of multispectral optoacoustic tomography necessitates improvements of the image quality available in real-time, as well as a reduction in the scanner financial cost. Deep learning approaches have recently unlocked the reconstruction of high-quality optoacoustic images in real-time. However, currently used deep neural network architectures require powerful graphics processing units to infer images at sufficiently high frame-rates, consequently greatly increasing the price tag. Herein we propose EfficientDeepMB, a relatively lightweight (17M parameters) network architecture achieving high frame-rates on medium-sized graphics cards with no noticeable downgrade in image quality. EfficientDeepMB is built upon DeepMB, a previously established deep learning framework to reconstruct high-quality images in real-time, and upon EfficientNet, a network architectures designed to operate of mobile devices. We demonstrate the performance of EfficientDeepMB in terms of reconstruction speed and accuracy using a large and diverse dataset of in vivo optoacoustic scans. EfficientDeepMB is about three to five times faster than DeepMB: deployed on a medium-sized NVIDIA RTX A2000 Ada, EfficientDeepMB reconstructs images at speeds enabling live image feedback (59 Hz) while DeepMB fails to meets the real-time inference threshold (14 Hz). The quantitative difference between the reconstruction accuracy of EfficientDeepMB and DeepMB is marginal (data residual norms of 0.1560 vs. 0.1487, mean absolute error of 0.642 vs. 0.745). There are no perceptible qualitative differences between images inferred with the two reconstruction methods.

关键词： Deep learning image quality In vivo Accuracy Translation Graphics processing units artificial neural networks Network architecture Real-time systems image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Adversarial Dual-Distortion image Rectification: Integrating Multi-Scale Discriminators and Central Self-Attention Modules

引用

IEEE Access 2025年 13卷 109172-109184页

作者： Jingfeng Luo Chengwan You Bochuan Zheng School of Electronic Information Engineering China West Normal University Sichuan China School of Computer Science China West Normal University Sichuan China Institute of Artificial Intelligence China West Normal University Sichuan China

Fisheye cameras, leveraging their wide-angle imaging advantages, have been widely adopted in industrial pipeline inspection, particularly for monitoring the inner wall conditions of cylindrical pipelines such as those in oil and gas wells. To achieve geometric reconstruction of pipeline wall images captured by fisheye lenses, this study proposes a method aimed at simultaneously correcting pipeline wall deformations (caused by cylindrical geometry) and fisheye lens distortions. Literature research reveals that fisheye distortion correction for pipeline wall images remains in its infancy, lacking real-scene datasets and data synthesis methods to support deep learning model training. Addressing this bottleneck, we constructed a dual-distortion mathematical model based on geometric optics principles and proposed a parametric simulation-based synthesis method to generate a specialized synthetic dataset tailored for training fisheye distortion correction networks. Existing fisheye correction methods exhibit performance degradation when processing dual-distortion images. To overcome this, we innovatively designed a central self-attention module adapted to fisheye imaging characteristics. This module dynamically enhances geometric feature extraction in the central region of distorted images, effectively suppressing distortion diffusion effects in peripheral areas. Additionally, we introduced a multi-scale discriminator with a pyramidal feature fusion architecture to strengthen the synergistic optimization of local textures and global structures in generative adversarial networks. Experimental results demonstrate that, compared to state-of-the-art correction methods, our approach achieves significant improvements in both objective quantitative metrics (e.g., SSIM, PSNR) and visual consistency. These findings validate the method’s robust capability to effectively correct fisheye-distorted pipeline wall images for industrial inspection applications, providing a reliable soluti

关键词： Distortion Pipelines Lenses Training Convolutional neural networks Cameras Feature extraction Calibration Oils Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

PLSAV: Parallel loop searching and verifying for loop closure detection

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2021年第5期15卷 683-698页

作者： Yang, Zhe Pan, Yun Deng, Lei Xie, Yuan Huan, Ruohong Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou Peoples R China Univ Calif Santa Barbara Dept Elect & Comp Engn Santa Barbara CA 93106 USA Zhejiang Univ Technol Coll Comp Sci & Technol Hangzhou Peoples R China

Visual simultaneous localization and mapping (vSLAM), one of the most important applications in autonomous vehicles and robots to estimate the position and pose using inexpensive visual sensors, suffers from error accumulation for long-term navigation without loop closure detection. Recently, deep neural networks (DNNs) are leveraged to achieve high accuracy for loop closure detection, however the execution time is much slower than those employing handcrafted visual features. In this paper, a parallel loop searching and verifying method for loop closure detection with both high accuracy and high speed, which combines two parallel tasks using handcrafted and DNN features, respectively, is proposed. A fast loop searching is proposed to link the bag-of-words features and histogram for higher accuracy, and it splits the images into multiple grids for high parallelism;meanwhile, a DNN feature extractor is utilized for further verification. A loop state control method based on a finite state machine to control these tasks is designed, wherein the loop closure detection is described as a context-related procedure. The framework is implemented on a real machine, and the top-2 best accuracy and fastest execution time of 80-543 frames per second (min: 1.84ms, and max: 12.45ms) are achieved on several public benchmarks compared with some existing algorithms.

关键词： image recognition Mobile robots Automata theory Computer vision and image processing techniques Machine learning (artificial intelligence) neural nets

来源：评论

学校读者我要写书评

暂无评论

LeQC-At: Learning Quantization Configurations During Adversarial Training for Robust Deep neural networks

LeQC-At: Learning Quantization Configurations During Adversa...

引用

Euromicro Symposium on Digital System Design

作者： Siddharth Gupta Salim Ullah Akash Kumar Chair of Embedded Systems Ruhr-Universität Bochum Bochum Germany

ISBN: (数字)9798350380385

ISBN: (纸本)9798350380392

Due to the high feature learning capability of Deep neural networks (DNNs), they are widely used in state-of-the-art machine learning tasks such as image and text recognition and natural language processing. However, in the recent developments of deep learning models, it has been observed that specially crafted inputs (adversarial samples) can deceive DNNs and result in incorrect predictions with high confidence. Such incorrect predictions by adversarially attacked DNNs can have devastating results in safety-critical applications. This situation can be further exacerbated in quantized DNNs, which employ reduced precision numbers to reduce the overall computational complexity of DNNs. To this end, this work proposes a framework for the joint optimization of DNNs to improve the natural accuracy of quantized DNNs and increase the robustness of the quantized models against adversarial attacks. In particular, the proposed framework employs quantization step size aware adversarial training of DNNs. Our proposed framework is generic and can be utilized with any quantization scheme that allows learning of quantization configurations during training. Furthermore, we present a novel loss function for adversarial training to improve the quantized networks' accuracy. For example, our 3-bit quantized adversarial training of ResNet-18, ResNet-34, and WideResNet shows up to 21.63%, 24.49%, and 15.08% higher inference accuracy with attacked data, respectively, when compared to 3-bit vanilla quantized adversarial training on benchmark datasets.

关键词： Training Representation learning Quantization (signal) Accuracy Text recognition Tiny machine learning artificial neural networks Predictive models Robustness Optimization

来源：评论

学校读者我要写书评

暂无评论

Intelligent Human Behaviour Detection Leveraging ConvLSTM and LRCN Deep Learning Approaches

Intelligent Human Behaviour Detection Leveraging ConvLSTM an...

引用

International Conference on Signal processing and Communication (ICSC)

作者： S Chakradhar Amingad Harika Janmanchi Mohammed Sultan K Shalini S Hariharan Bhanu Prakash Maddala Artificial Intelligence and Data Science Vardhaman College of Engineering Hyderabad Telangana India Artificial Intelligence and Data Science Vardhaman college of engineering Hyderabad India

ISBN: (数字)9798331511890

ISBN: (纸本)9798331511906

This research delves into intelligent human behavior detection from video sequences by employing advanced deep learning methodologies. The study harnesses the capabilities of Convolutional Long Short-Term Memory (ConvLSTM) and Long-term Recurrent Convolutional networks (LRCN) architectures, which are adept at capturing both temporal and spatial dynamics in video data. Using the extensive Kinetics dataset, which includes a wide range of human activities and contexts, the video data was converted into sequential image frames for model training. The ConvLSTM and LRCN models were implemented and trained with the TensorFlow/Keras framework on GPU-accelerated systems. During training, both models exhibited effective learning, with improved accuracy and reduced loss. However, validation metrics revealed fluctuations, indicating possible overfitting and challenges in generalizing to new data. To address these issues, regularization techniques such as dropout and early stopping were applied, alongside comprehensive hyperparameter tuning. Additionally, the study investigated the use of transfer learning to enhance model performance. This research lays the groundwork for sophisticated video-based human behavior detection systems, with applications in surveillance, healthcare, and human-computer interaction. Future work will aim to bolster model robustness and broaden the scope of action recognition.

关键词： Training Deep learning Surveillance Video sequences Transfer learning Robustness Behavioral sciences Convolutional neural networks Long short term memory Tuning

来源：评论

学校读者我要写书评

暂无评论

Skin cancer classification using Convolutional neural networks 11

Skin cancer classification using Convolutional neural networ...

引用

11th International Conference on Cloud Computing, Data Science and Engineering (Confluence)

作者： Subramanian, R. Raja Achuth, Dintakurthi Kumar, P. Shiridi Reddy, Kovvuru Naveen Kumar Amara, Srikar Chowdary, Adusumalli Suchan Kalasalingam Acad Res & Educ Dept Comp Sci & Engn Virudunagar Tamil Nadu India

ISBN: (纸本)9781665414517

There is a necessary need for early detection of skin cancer and can prevent further spread in some cases of skin cancers, such as melanoma and focal cell carcinoma. Anyhow there are several factors that have bad impacts on the detection accuracy. In Recent times, the use of image processing and machine vision in the field of healthcare and medical applications is increasing at a greater phase. In this paper, we are using the Convolution neural networks to detect and classify the class of cancer based on historical data of clinical images using *** of our objectives through this research are,to build a CNN model to detect skin cancer with an accuracy of >80%,to keep the false negativity rate in the prediction to below 10%, to reach the precision of above 80% and do visualization on our Data. Simulation results show that the proposed method has superiority towards the other compared methods.

关键词： Skin cancer diagnosis Deep learning Convolutional neural networks [Standard CNN],Skin cancer artificial neural networks

来源：评论

学校读者我要写书评

暂无评论

Efficient Detection of Alzheimer's Disease on Early Stages Using Deep Learning on MRI Data

Efficient Detection of Alzheimer's Disease on Early Stages U...

引用

artificial Intelligence and Machine Learning applications Theme: Healthcare and Internet of Things (AIMLA), International Conference on

作者： B. Neelambaram D. Ganesh Vishnu Vandana Devi V Dinesh P D.Shobha Rani M. Sunil Kumar Siddhartha Academy of Higher Education deemed to be University Vijayawada Andhra Pradesh Mohan Babu University (Erstwhile Sree Vidyanikethan Engineering College(Autonomous) Tirupati Andhra Pradesh India Department of English Siddhartha Academy of Higher Education Deemed to be University Vijayawada Department of MCA Mohan Babu University (Erstwhile Sree Vidyanikethan Engineering College(Autonomous)) Tirupati Andhra Pradesh Department of Computer Science and Engineering Chadalawada Ramanamma Engineering College Tirupathi Andhra Pradesh India

ISBN: (数字)9798331538965

ISBN: (纸本)9798331538972

Alzheimer's disease (AD) is a human brain disorder mainly affecting the older persons, which could be leads to cognitive decline and memory loss. The detection of Alzheimer's considered the important because of its slow disease progression and checking high quality of the human brain image for patients. This main aim of this work is detecting the human brain disease using deep learning techniques on Magnetic Resonance Imaging (MRI) data for the processing the brain and display the results is whether the person is affected the disease or normal condition in the brain. The MRI scans detailed information on the brain structures is used to spot the subtle changes that can be occur, once a person receives an onset of Alzheimer's Disease. This work involves preprocessing of MRI data to process the image quality and remove noise, followed by feature extraction to capture critical patterns which are main indicative of the disease. The Convolutional neural networks (CNNs) are applied to the MRI images for analyzing to use their ability to automatically detect the complex patterns in high-dimensional data. Various architectures are trained and evaluated for performance metrics such as accuracy, sensitivity, and specificity to classify most effective model in order to detect the Alzheimer's. The comparative outcomes establish the possible of deep learning to outperform traditional diagnostic methods by offering faster, more accurate, and reliable predictions. This work is used to the field of medical imaging and significant suggestions for improving Alzheimer's diagnosis and advancing accurate medicine.

关键词： Deep learning Accuracy Sensitivity Magnetic resonance imaging Medical services Sensitivity and specificity Feature extraction Brain modeling Convolutional neural networks Alzheimer's disease

来源：评论

学校读者我要写书评

暂无评论

Deepfake image and Video Detection Using Deep Learning Algorithms

Deepfake Image and Video Detection Using Deep Learning Algor...

引用

Multi-Agent Systems for Collaborative Intelligence (ICMSCI), International Conference on

作者： Srivanth Srinivasan Nischitha P Akshita Chavan Mohana Artificial Intelligence & Machine Learning Engineering RV College of Engineering® Bengaluru India Computer Science & Engineering (Cyber Security) RV College of Engineering® Bengaluru India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509835

Deep learning (DL) algorithms are swiftly finding applications in computer vision and natural language processing. Nonetheless, they can also be employed for creating convincing deepfakes, which are challenging to distinguish from reality. The advancements in image and video technology and tools, especially on social media platforms, potentially lead to misuse for malicious purposes like blackmail or defamation. To tackle this issue, several group of researchers tried upon spreading or creating awareness on real or fake data. The proposed approach involves combining Deepfake generation using GANs and Autoencoders with a Deepfake detection method. The aim of this initiative is exclusively to combat disinformation and online fraud for the welfare of the general population. Deepfakes, products of AI, have become increasingly realistic, rendering it nearly difficult to distinguish the content. Auto-encoders with sufficient time can achieve about 92 % accuracy. As the generator improves, the discriminator performance worsens as it struggles to differentiate real or fake data. A perfect generator results in 50% accuracy. With advancements in computational capacity and data availability, the proposed DDM (Deepfake Detection Model) has achieved greater accuracy rate of up to 92.3%.

关键词： Deep learning Deepfakes Accuracy Social networking (online) Streaming media Feature extraction Generators Convolutional neural networks artificial intelligence Research and development

来源：评论

学校读者我要写书评

暂无评论

Automatic Detection of Leukemia through Convolutional neural Network

Automatic Detection of Leukemia through Convolutional Neural...

引用

2022 International Conference on Frontiers of Information Technology, FIT 2022

作者： Arif, Rida Akbar, Shahzad Farooq, Ahmad Bilal Ale Hassan, Syed Gull, Sahar Riphah International University Department of Computing Pakistan

ISBN: (纸本)9798350345933

Leukemia is a fatal cancer disease that develops in blood-forming tissue by the excessive development of white blood cells (WBCs) in the human body. However, a bone marrow test is recommended by the pathologist to diagnose leukemia and further types of leukemia. Leukemia has two classes i.e., acute and chronic leukemia. Therefore, early leukemias detection enables preventative actions to be taken to avoid any harm to human life. In addition, several manual and automatic methods have been proposed, however, they possess some drawbacks and are inefficient for the precise detection of leukemia. This research proposes a deep learning-based framework for precise and automatic leukemia identification using microscopic images. The proposed framework comprises four stages which are pre-processing, data augmentation, segmentation, and the classification of leukemia. Moreover, pre-processing is utilized to clean the dataset images and eliminate the noise. Following that, data augmentation approaches have been employed to increase the number of images, and remove the class imbalance, and overfitting problems. The modified Convolutional neural Network (CNN) based model is employed to segment the leukemia images. A well-known pre-trained AlexNet architecture has been used for classification. Besides that, a publicly available dataset Acute Lymphoblastic Leukemia image DataBase (ALL-IDB) has been utilized to train and test the proposed model. The proposed model yielded 98.05% accuracy, a specificity of 97.59%, 100% of recall, and a 99.06% of F1-score. The experimentation results demonstrate that this model is effective and reliable for leukemia identification using the ALL-IDB dataset and suitable for deployment in clinical applications. © 2022 IEEE.

关键词： Convolutional neural networks

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：