检索结果-内蒙古大学图书馆

International Automatic Control Conference

作者： Jih-Chieh Yeh Yu-Ping Liao Electrical Engineering Department Chung Yuan Christian University R.O.C Taoyuan Taiwan

ISBN: (数字)9798350354904

ISBN: (纸本)9798350354911

Recently, with the development of the artificial intelligence (AI). The application of Deep Learning is expending into different field. image recognition, personalized recommendations in search engines, Natural Language processing (NLP) and Speech Recognition are based on the Deep Learning. The continued development and integration of these applications can also lead to a more convenient lifestyle. Including Autonomous Vehicle, Facial Recognition and more. Based on the structure of Deep Learning. We transfer the data of the ambient sound into the Log-Mel spectrogram. Then using the Log-Mel spectrogram to train the Convolutional neural Network (CNN) model. After this, we deployed the trained model onto the STM32 SensorTile board for experimentation.

关键词： Training Deep learning Accuracy Roads Feature extraction Natural language processing Convolutional neural networks Spectrogram Testing Sound recognition

来源：评论

学校读者我要写书评

暂无评论

Advancements in deep structured learning based medical image interpretation

引用

JOURNAL OF INFORMATION & OPTIMIZATION SCIENCES 2022年第5期43卷 1131-1138页

作者： Pasupathy, Vijayalakshmi Khilar, Rashmita Panimalar Engn Coll Dept Comp Sci & Engn Chennai Tamil Nadu India Saveetha Inst Med & Tech Sci Dept Informat Technol Saveetha Sch Engn Chennai Tamil Nadu India

Deep learning is a fast flattering highest development in the machine learning technique leading to augment performance in diverse medical applications. The tributary artificial intelligence is rapidly becoming metamorphic in the domain of Healthcare Technology rendering facility to analyze the medical data with high rapidity and exactness. Due to the overwhelming development of artificial intelligence (Al), the contemporary encroachments of deep learning employing advanced and futuristic deep learning methods has become a vivacious research domain for processing medical data in the health care industries. this paper presents a complete analysis about the deep structured learning based methods used in medical image analysis and also elucidate about the advancement of convolutional neural network which is a major dass of deep neural networks used in medical image processing. It also gives insight about the fundamentals of deep learning methods and reviews in image interpretation. It also explains about the implementation of an integrated framework for medical image synthesis combining Speckle GAN and Di-Comr-AE-Net.

关键词： Deep belief networks (DBNs) Denoising autoencoder (DAE) Long short-term memory (LSTM) Recurrent neural network (RNN) Dilated convolution autoencoder network(Di-Conv-AE-Net)

来源：评论

学校读者我要写书评

暂无评论

image Segmentation and Recognition of Oracle Bone Topographies Based on Deep Learning

Image Segmentation and Recognition of Oracle Bone Topographi...

引用

image processing and Computer applications (ICIPCA), IEEE International Conference on

作者： Zhengen Lv Yihui Liu Yunjie Cai Qi Liu College of Mechanical and Control Engineering Guilin University of Technology Guilin China School of Mechanical and Electronic Engineering Jingdezhen Ceramic University Jingdezhen China College of Mechanical and Electrical Engineering Guangxi Agricultural Engineering Vocational Technical College Chongzuo China School of Energy and Electrical Engineering Qinghai University Xining China

ISBN: (数字)9798350360240

ISBN: (纸本)9798350384161

image segmentation and recognition of oracle bone topography is a common challenge for in-depth study of oracle bones. The key lies in how to accurately and efficiently process and recognize the original oracle bone topography images and solve the practical problems in the research based on the image digitization results. The research of this topic helps to promote the research of oracle bone holography and the construction of digitization project. image segmentation of oracle bone tablet is to use digital image processing and computer vision technology to extract independent text regions with distinctive image features and non-overlapping each other in the complex background of the original topographic image. The study of automatic segmentation and recognition of oracle bone characters on original topos has important social significance. In this paper, the image preprocessing model based on the channel attention mechanism is firstly utilized to initially judge and identify the interference elements present in the original oracle bone topography image. Meanwhile, a joint filtering noise reduction model is established to effectively remove the location-like noise and artificial texture in the image. Then, a fast oracle bone image segmentation model based on neural network is established, and each single character region in the image is successfully segmented out by building and training the model through deep learning. In this paper, 76 common oracle scripts and their variant scripts are trained and recognized in combination with the oracle recognition enhancement model, and a one-to-many link between Chinese characters and oracle images is established. The accuracy and efficiency of oracle bone recognition are further improved by segmenting and classifying the test set images for recognition.

关键词： Deep learning image segmentation image recognition Accuracy Computational modeling Noise reduction neural networks Surfaces Bones Character recognition

来源：评论

学校读者我要写书评

暂无评论

Deep Complex-valued Radial Basis Function neural networks and Parameter Selection

Deep Complex-valued Radial Basis Function Neural Networks an...

引用

International Symposium on Wireless Communication Systems (ISWCS)

作者： Jonathan A. Soares Vinícius H. Luiz Dalton S. Arantes Kayol S. Mayer Department of Communications School of Electrical and Computer Engineering Universidade Estadual de Campinas – UNICAMP Campinas SP Brazil

ISBN: (数字)9798350362510

ISBN: (纸本)9798350362527

In the ever-evolving field of artificial neural networks and learning systems, complex-valued neural networks (CVNNs) have become a cornerstone, achieving exceptional performance in image processing and telecommunications. More precisely, in digital communication systems, CVNNs have been delivering significant results in tasks like equalization, channel estimation, beamforming, and decoding. Among the CVNN architectures, the complex-valued radial basis function neural network (C-RBF) stands out, especially when operating in noisy environments such as 5G multiple-input multiple-output (MIMO) systems. In such a context, this paper extends the classical shallow C-RBF to deep architectures, increasing its flexibility for a wider range of applications. Also, based on the parameter selection of the phase transmittance radial basis function (PT-RBF) neural network, we propose an initialization scheme for the deep C-RBF. Via rigorous simulations conforming to 3GPP TS 38 standards for digital communications, our method not only outperforms conventional initialization strategies like random, K-means, and constellation-based methods but it also seems to be the only approach to achieve successful convergence for deep C-RBF architectures. These findings pave the way to more robust and efficient neural network deployments in complex-valued digital communication systems.

关键词： 5G mobile communication Neurons Radial basis function networks Digital communication Robustness 3GPP Biological neural networks

来源：评论

学校读者我要写书评

暂无评论

Tuning neural networks for Geofencing applications

Tuning Neural Networks for Geofencing Applications

引用

2020 International Conference on Robotics and artificial Intelligence, RoAI 2020

作者： Eldho John, George Joseph Mamutil, Rohit Danlaw Inc Bangalore India Volvo Group Bangalore India

neural networks have been used in a wide range of applications such as image processing, Object detection, Weather and Economic Forecast and many other areas where algorithms which could learn from data have been revolutionizing. This study elaborates on the usage of neural networks in the field of geofencing. Ever since the advent of satellite-based Maps, Geofencing has been mentioned as one of the key possible features. With the booming usage of smartphones and IoT based devices, Geofence has a key role to play in location based advertising, marketing, GPS tracking, etc. This paper explains how to use neural networks for detecting whether a point is in geofence or not. It is proved in this study that neural networks perform better over commonly used techniques such as the Ray casting method in some scenarios. © Published under licence by IOP Publishing Ltd.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

PCUNet: A Context-Aware Deep Network for Coarse-to-Fine Point Cloud Completion

引用

IEEE SENSORS JOURNAL 2022年第15期22卷 15098-15110页

作者： Zhao, Meihua Xiong, Gang Zhou, Mengchu Shen, Zhen Liu, Sheng Han, Yunjun Wang, Fei-Yue Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Chinese Acad Sci Beijing Engn Res Ctr Intelligent Syst & Technol Inst Automat Beijing 100190 Peoples R China Chinese Acad Sci Cloud Comp Ctr Guangdong Engn Res Ctr 3D Printing & Intelligent Dongguan 523808 Peoples R China New Jersey Inst Technol Helen & John C Hartmann Dept Elect & Comp Engn Newark NJ 07102 USA Qingdao Acad Intelligent Ind Intelligent Mfg Ctr Qingdao 266109 Peoples R China

Point cloud completion aims at predicting a complete 3D shape from an incomplete input. It has important applications in the fields of intelligent manufacturing, augmented reality, virtual reality, self-driving cars, and intelligent robotics. Although deep learning-based point cloud completion technology has developed rapidly in recent years, there are still unsolved problems. Previous approaches predict each point independently and ignore contextual information. And, they usually predict a complete 3D shape based on a global feature vector extracted from an incomplete input, which leads to missing of some fine-grained details. In this paper, motivated by the transposed convolution and the "UNet" structure in neural networks for image processing, we propose a context-aware deep network termed as PCUNet for coarse-to-fine point cloud completion. It adopts an encoder-decoder structure, in which the encoder follows the design of the relation-shape convolutional neural network (RS-CNN), and the decoder consists of fully-connected layers and two stacked decoder modules for predicting complete point clouds. The contributions are twofold. First, we design the decoder module as a coordinate-guided context-aware upsampling module, in which contextual information can be taken into full account by neighbor aggregation. Second, to preserve fine-grained details in the input, we propose attention-enhanced skip connections for effective information propagation from the encoder to the decoder. Experiments are conducted on the widely used PCN and KITTI datasets. The results show that our proposed approach achieves competitive performance compared to the existing state-of-the-art approaches in terms of the Chamfer distance and the computational complexity metrics.

关键词： Point cloud compression Three-dimensional displays Decoding Shape Feature extraction Sensors neural networks Point cloud completion deep learning context-aware attention-enhanced skip connections

来源：评论

学校读者我要写书评

暂无评论

A New Lightweight Hybrid Model for Pistachio Classification Using Transformers and EfficientNet

引用

IEEE Access 2025年 13卷 85857-85872页

作者： Cakmak, Muhammet Sinop University Department of Computer Engineering Sinop57100 Turkey

In recent years, Vision Transformers (ViTs) have gained prominence as a highly effective method for image classification, often outperforming traditional Convolutional neural networks (CNNs). However, their relatively slow processing speed limits their practical use, particularly in real-time applications. Conversely, CNN-based transfer learning models provide faster inference but may struggle with classification accuracy on complex datasets. To address these challenges, Temporal Coordinate Attention (TCA) modules have been introduced to optimize efficiency and performance. This study proposes a hybrid architecture combining EfficientNet, Vision Transformer, and Temporal Channel Attention modules that integrates the accuracy of ViTs, the computational efficiency of CNNs, and the enhancement capabilities of TCA modules. The model is designed to classify Siirt and Kirmizi pistachio varieties with high precision. It achieves outstanding results, including 99.07% accuracy, 99.12% recall, and a Cohen’s Kappa score of 98.10%. These findings highlight the model’s robustness, demonstrating its ability to perform reliable classifications with minimal bias, making it well-suited for real-world applications. © 2013 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Automated Detection of Tomato Leaf Diseases Using Convolutional neural networks

Automated Detection of Tomato Leaf Diseases Using Convolutio...

引用

Automation and Computation (AUTOCOM), International Conference on

作者： Hemant Kumar Yadav Varsha Srivastava Sreenu Banoth Rohini Daksh Rawat Priyansha Rawat Artificial Intelligence & Data Science IIMT College of Engineering Greater Noida Uttar Pradesh India School of Computer Science and Engineering IILM University Greater Noida Uttar Pradesh Department of Computer Science and Engineering Chandigarh Engineering College Chandigarh Group of Colleges Mohali Punjab India Computer Science and Engineering Graphic Era Hill University Dehradun India

ISBN: (数字)9798331542375

ISBN: (纸本)9798331542382

In this study, an automated system to detect tomato leaf diseases with Convolutional neural networks (CNNs) is introduced to aid farmers in early disease identification to avert crop losses and promote yield. The dataset is multiple classes of tomato leaf diseases, and the model was trained and tuned extensively for its performance. In particular, a high training accuracy of 95.73%, and validation accuracy of 89.85% was achieved by CNN. Furthermore, the system's precision, recall, and F1 metric were around 90%, indicating the balanced system performance. The proposed CNN model is shown to be more reliable and faster than the traditional image processing techniques and hence is an appropriate choice for real time agricultural applications. The integration of modern varieties technologies in AI based approaches in agriculture which will feed productivity while maintaining Sustainability. The results indicate that the proposed system could provide an effective means for disease detection for farmers and may alleviate some of the need for chemical treatments.

关键词： Training Productivity Plant diseases Accuracy Crops Predictive models Agriculture Convolutional neural networks Monitoring Diseases

来源：评论

学校读者我要写书评

暂无评论

Classification of MRI's images using the Fractional Fourier Transform and the Local Binary Pattern information by artificial neural Network 29

Classification of MRI's images using the Fractional Fourier ...

引用

Conference on Infrared Remote Sensing and Instrumentation XXIX

作者： Estrada Marmolejo, Carlos A. Alvarez-Borrego, Josue Garcia-Torales, Guillermo Flores Nunez, Jorge Luis Univ Guadalajara Av Revoluc 1500 Guadalajara 44430 Jalisco Mexico CICESE Carretera Ensenada Tijuana 3918 Zona Playitas Ensenada 22860 Baja California Mexico

ISBN: (数字)9781510644991

ISBN: (纸本)9781510644991;9781510644984

Pattern recognition techniques are widely used in computer vision, classification of radio signals, and voice recognition. The fractional Fourier transform is used to recognize patterns using binary rings masks and segment images. This technique has the characteristic of being invariant to position and rotation and finally obtaining a one-dimensional signature. On the other hand, neural networks are used for pattern recognition based on a deep neural network algorithm. It has the characteristic of training large datasets with millions of images. artificial neural networks(ANNs) are used for several applications such as pattern recognition and classification of input data. In particular, the ANN has been used to evaluate medical images from the brain to assess if the image corresponds to Alzheimer's disease. One disadvantage of the neural network is a large amount of time to learn depending on the number of patterns to be identified or classified and the ability to adapt and recognize patterns. Besides, the fractional Fourier transform cannot analyze a large amount of information. In this work, a comparison between the artificial neural Network and the Fractional Fourier Transform is presented to determine which will be the best for recognizing a batch of selected medical images. We propose a reconstruction method using both techniques for precise image recognition and the evaluation of their respective metrics such as accuracy, precision, sensitivity, and specificity. The medical images regarding Alzheimer's disease are no dementia, very mild dementia, mild dementia presenting the best perfomance regarding the receiver operating characteristics and moderate dementia was the worst classified related to the number of images of the dataset.

关键词： Pattern recognition one-dimensional signature fractional Fourier transform binary rings masks artificial neural Network computer vision image processing local binary pattern neural network

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Ternary Multiplier

Energy-Efficient Ternary Multiplier

引用

IEEE International Conference on artificial Intelligence Circuits and Systems (AICAS)

作者： Rishi Agrawal Narayanabhatla Savyasachi Abhijith Uppugunduru Anil Kumar Sreehari Veeramachaneni Syed Ershad Ahmed Department of Electrical and Electronics Engineering BITS Pilani Hyderabad Telangana India Department of Electronics and Communication Engineering Faculty of Science and Technology (IcfaiTech) The ICFAI Foundation for Higher Education (Deemed to be University) Hyderabad Telangana India Department of Electronics and Communication Engineering GRIET Hyderabad Telangana India

ISBN: (数字)9798350383638

ISBN: (纸本)9798350383645

Multiplication is a fundamental arithmetic operation with applications in various fields, such as neural networks, signal and image processing. Ternary multiplication, an emerging area of research, explores the multiplication of numbers represented in the ternary numeral system. This paper proposes a new technique for computing partial product trits of multi-digit multiplication, along with improved adder modules, intending to reduce power dissipation. Two parallel paths have been proposed to calculate the final product, thereby improving the latency of the multiplier. The internal adder modules of the multiplier have been designed to optimise the number of transistors, resulting in lower power consumption. Results prove that the proposed design significantly improves power, delay, and the power-delay product (PDP). The proposed three-trit multiplier shows a 20.7% improvement in power, 23.17% in delay, 39.11% in PDP, and 18.76% in transistor count compared to the existing design.

关键词： Power demand Circuits and systems image processing neural networks Energy efficiency Delays Power dissipation

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：