检索结果-内蒙古大学图书馆

30th National and 8th International Iranian Conference on Biomedical Engineering (ICBME)

作者： Mirani, Mehri Soltani-Gol, Mohammad Soltanian-Zadeh, Hamid Univ Tehran Sch Elect & Comp Engn CICPE Tehran Iran

ISBN: (纸本)9798350359732;9798350359749

Brain tumor segmentation is essential in medical imaging for accurate diagnosis and effective treatment planning. Brain tumors are formed during the uncontrolled growth of cells and are divided into two primary and secondary categories. The most common type of primary brain tumors is gliomas, which are examined in this research. In recent years, deep learning models have shown promising results in automating tumor segmentation from medical imaging data. This study compares the performance of Variational Autoencoder (VAE) and Adversarial Autoencoder (AAE) networks for brain tumor segmentation based on an encoder-decoder architecture. The encoder part extracts image features, while the decoder part reconstructs segmentation labels. These two networks are powerful generative models that capture the underlying distribution of the data. This capability makes them suitable for extracting meaningful tumor features. In the VAE network, a variational autoencoder branch is added to reconstruct the input images. In the adversarial network, the discriminator part is added to train the encoder-decode network and helps the decoder to generate outputs that closely match the tumor labels. The results indicate that both VAE and AAE achieve satisfactory segmentation accuracy. However, AAE achieves higher Dice scores compared to VAE, indicating that it is more effective at segmenting tumor regions.

关键词： Brain tumor segmentation deep learning encoder-decoder Variational Autoencoder (VAE) Adversarial Autoencoder (AAE)

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Model for Multilingual OCR 17th

A Hybrid Model for Multilingual OCR

引用

17th International Conference on Document Analysis and Recognition (ICDAR)

作者： Etter, David Carpenter, Cameron King, Nolan Johns Hopkins Univ Human Language Technol Ctr Excellence Baltimore MD 21218 USA SCALE Laurel MD USA

ISBN: (纸本)9783031416750;9783031416767

Large-scale document processing pipelines are required to recognize text in many different languages. The writing systems for these languages cover a diverse set of scripts, such as the standard Latin characters, the logograms of Chinese, and the cursive right-to-left of Arabic. Multilingual OCR continues to be a challenging task for document processing due to the large vocabulary sizes and diversity of scripts. This work introduces a multilingual model that recognizes over nine thousand unique characters and seamlessly switches between ten different scripts. Our transformer-based encoder-decoder approach combines a CTC objective on the encoder with a cross-entropy objective on the full autoregressive decoder. The hybrid approach allows the fast non-autoregressive encoder to be used in standalone mode or with the full autoregressive decoder. We evaluate our approach on a large multilingual dataset, where we achieve state-of-the-art character error rate results in all thirteen languages. We also extend the encoder with auxiliary heads to identify language, predict font, and detect vertical lines.

关键词： Multilingual OCR Transformer Hybrid encoder-decoder CTC Auxiliary heads Synthetic data

来源：评论

学校读者我要写书评

暂无评论

Multi-task Learning Based Skin Segmentation 16th

Multi-task Learning Based Skin Segmentation

引用

16th International Conference on Knowledge Science, Engineering and Management (KSEM)

作者： Tan, Taizhe Shan, Zhenghao Guangdong Univ Technol Sch Comp Sci & Technol Guangzhou 510006 Peoples R China Heyuan Bay Area Digital Econ Technol Innovat Ctr Heyuan 517001 Peoples R China

ISBN: (纸本)9783031402883;9783031402890

Skin segmentation is a critical task in computer vision that has diverse applications in several fields such as biometrics, medical imaging, and video surveillance. Despite its importance, the acquisition of high-quality data remains a significant challenge in skin segmentation research. In this paper, we propose a novel skin segmentation algorithm for single-person images by utilizing a dual-task neural network built on the multi-task learning framework. Specifically, the algorithm employs an encoder-decoder architecture consisting of a shared backbone, two dynamic encoders, and a decoder. The dynamic encoders use dynamic convolution to extract more spatial location information, while the decoder utilizes a query-based dual-task approach that allows each task to utilize the information generated by the other one efficiently. The experimental results indicate that the proposed skin segmentation algorithm outperforms or matches the current state-of-the-art techniques on the benchmark test set.

关键词： Skin segmentation query-based multi-task learning encoder-decoder deep learning

来源：评论

学校读者我要写书评

暂无评论

The Application of Generating API Call Sequence Code for Android Driven by Neural Network

The Application of Generating API Call Sequence Code for And...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Yang, Jingbo Wu, Wenjun Ren, Jian Beihang Univ Sch Comp Sci & Engn Beijing Peoples R China Beihang Univ Inst Artificial Intelligence Beijing Peoples R China

ISBN: (纸本)9781665488679

API, namely application programming interface, can help developers implement their functions conveniently. To implement a function like dialing in Android, developers sometimes need to call many APIs organised in a special pattern, called API call sequence. However, existing methods rarely focus on code generation for API call sequence. In this paper, we introduce neural network into the application field of generating API call sequence code for Android. The purpose is realising Android code automatically generation by inputting function description. To reach this goal, we first design an API call sequence code graph which is named ACSCG to well represent the Android function code structure and then we convert the ACSCG to API call sequence. Besides, we devise an AI model based on encoder-decoder neural network to study the corresponding relation feature of function description and API call sequence. When finishing training the model, one can give a function description to it and generate corresponding API call sequence. After all above has been done, an algorithm is implemented to successfully convert the API call sequence into target code. To verify the efficiency of our method, we collect high quality code from Github and Gitee to build a dataset including 1000 items associated with essential functions in Android such as taking photo, file management, android browser and so on. The experiment shows that our model has a better performance in generating API call sequence code than state-of-the-art technologies.

关键词： API call sequence neural network encoder-decoder AI code generation

来源：评论

学校读者我要写书评

暂无评论

Bridging the Language Gap: Enhancing English-to-Telugu Translation using NMT and Encoding Decoding Techniques 15

Bridging the Language Gap: Enhancing English-to-Telugu Trans...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Asmitha, M. Kavitha, C.R. Amrita Vishwa Vidyapeetham Amrita School of Computing Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798350370249

Language translation, particularly in the context of natural language processing, involves converting text or speech from one language into another while preserving its meaning, context, and cultural nuances. This task is critical for enabling communication across different languages and is a significant area of research and development in artificial intelligence. Machine translation has a wide range of real-time applications that have significantly impacted various fields by bridging language barriers. Some of the real time application includes Google Translate, Microsoft Translator, Skype Translator. NMT has revolutionized language translation with models like Google's Transformer, OpenAI's GPT, Facebook's BART, and Google's T5. Development of a machine translation system to translate English sentences into Telugu. The system utilizes a neural network model based on an encoder-decoder architecture with attention mechanism. Preprocessing techniques are applied to the dataset, including lowercasing, removal of special characters, and tokenization. The sparse categorical cross-entropy loss function and Adam optimizer are used to train the model. The actual vs. projected output is used to evaluate performance. Pre-trained GloVe word embeddings are used in the implementation to improve the model's capacity to extract semantic information. © 2024 IEEE.

关键词： Attention mechanism encoder-decoder GloVe word embeddings Machine translation Neural network Pre-processing

来源：评论

学校读者我要写书评

暂无评论

Transformer-Based Multivariate Time Series Forecasting

Transformer-Based Multivariate Time Series Forecasting

引用

2024 IEEE International Students' Conference on Electrical, Electronics and Computer Science, SCEECS 2024

作者： Bharti, Mukesh Kumar Wadhvani, Rajesh Gyanchandani, Manasi Gupta, Muktesh Maulana Azad National Institute of Technology Department of Cse Bhopal India

ISBN: (纸本)9798350348460

This study explores multivariate time series forecasting, centering on the transformer model. It examines the shortcomings of other predictive models like Recurrent Neural Networks (RNN) and Temporal Convolutional Networks (TCN), particularly their inadequacies in handling autocorrelation. The transformer model stands out for its accuracy, thanks to its attention mechanism that focuses on essential parts of the input. The research introduces a novel approach that employs the transformer's architecture for effective feature selection in time series data. A vital aspect of this approach is the use of unsupervised pre-training, which shows superior results compared to traditional fully supervised methods. This advancement underscores the effectiveness of unsupervised learning in time series regression, offering significant benefits for diverse scientific and industrial fields. © 2024 IEEE.

关键词： Attention Mechanism Autocorrelation Deep Learning encoder-decoder Heteroskedasticity Quantile Regression Time Series Forecasting Transformers

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Generative Neural Attention-based Service Chatbot

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS 2022年第8期13卷 742-751页

作者： Suhaili, Sinarwati Mohamad Salim, Naomie Jambli, Mohamad Nazim Pre Univ Kota Samarahan Sarawak Malaysia Univ Teknol Malaysia Fac Comp Skudai 81310 Johor Malaysia Univ Teknol Malaysia Ibnu Sina Inst Sci & Ind Res UTM Big Data Ctr Skudai 81310 Johor Malaysia Univ Malaysia Sarawak Fac Comp Sci & Informat Technol Kota Samarahan Sarawak Malaysia

Companies constantly rely on customer support to deliver pre-and post-sale services to their clients through websites, mobile devices or social media platforms such as Twitter. In assisting customers, companies employ virtual service agents (chatbots) to provide support via communication devices. The primary focus is to automate the generation of conversational chat between a computer and a human by constructing vir-tual service agents that can predict appropriate and automatic responses to customers' queries. This paper aims to present and implement a seq2seq-based learning task model based on encoder-decoder architectural solutions by training generative chatbots on customer support Twitter datasets. The model is based on deep Recurrent Neural Networks (RNNs) structures which are uni-directional and bi-directional encoder types of Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU). The RNNs are augmented with an attention layer to focus on important information between input and output sequences. Word level embedding such as Word2Vec, GloVe, and FastText are employed as input to the model. Incorporating the base architecture, a comparative analysis is applied where baseline models are compared with and without the use of attention as well as different types of input embedding for each experi-ment. Bilingual Evaluation Understudy (BLEU) was employed to evaluate the model's performance. Results revealed that while biLSTM performs better with Glove, biGRU operates better with FastText. Thus, the finding significantly indicated that the attention-based, bi-directional RNNs (LSTM or GRU) model significantly outperformed baseline approaches in their BLEU score as a promising use in future works.

关键词： Sequence-to-sequence encoder-decoder service chatbot attention-based encoder-decoder Recurrent Neural Network (RNN) Long Short-Term Memory (LSTM) Gated Recurrent Unit (GRU) word embedding

来源：评论

学校读者我要写书评

暂无评论

Dense spatially-weighted attentive residual-haze network for image dehazing

引用

APPLIED INTELLIGENCE 2022年第12期52卷 13855-13869页

作者： Singh, Mohit Laxmi, Vijay Faruki, Parvez MNIT Jaipur Rajasthan India AVPTI Rajkot Gujarat India

Haze severely affects computer vision algorithms by degrading the quality of captured images and results in image data loss. With several available approaches for dehazing, single image dehazing is most preferred and challenging. We proposed a Dense Spatially-weighted Attentive Residual-haze Network (DSA Net), a novel end-to-end encoder-decoder architecture to learn the residual haze layer between the hazy and haze-free image. We use encoder-decoder blocks with multiple skip connections to improve feature propagation. Feature Learning block uses a novel Residual Inception fused with Attention (RIA) block to learn the complex non-linearity from features extracted from the encoder part. Learning residual image is more straightforward than the whole haze-free image, and it improves the ability of the network to estimate the haze thickness accurately. DSA Net learns this less complex residual-map from the hazy input image and subtracts it from the input to obtain the dehazed images. Detail ablation study shows the effectiveness of different modules used in our architecture. Experiment results on different haze conditions demonstrate that our method shows significant improvement over other state-of-the-art methods.

关键词： Dehazing Attention mechanism encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

FMAM-Net: Fusion Multi-Scale Attention Mechanism Network for Building Segmentation in Remote Sensing Images

引用

IEEE ACCESS 2022年 10卷 134241-134251页

作者： Ye, Huanran Zhou, Run Wang, Jianhao Huang, Zhiliang Yiwu Ind & Commercial Coll Sch Mech & Elect Informat Jinhua 322000 Zhejiang Peoples R China Zhejiang Normal Univ Coll Phys & Elect Informat Engn Jinhua 321004 Zhejiang Peoples R China

As the largest target in remote sensing images, buildings have important application value in urban planning and old city reconstruction. However, most networks have poor recognition ability on high resolution images, resulting in blurred boundaries in the segmented building maps. Then, the similarity between buildings and backgrounds will lead to inter-class indistinction. Finally, the diversity of buildings brings difficulties to segmentation, which requires the network to have better generalization ability. To address these problems, we propose Fusion Multi-scale Attention Mechanism Network (FMAM-Net). Firstly, we design Feature Refine Compensation Module(FRCM) to improve the boundary ambiguity problem, including Feature Refinement Module(FRM) and Feature Compensation Module(FCM). FRM utilizes the densely connected architecture to refine features and increase recognition capabilities. FCM introduces low-level features to make up for the lack of boundary information in high-level features. Secondly, to handle inter-class indistinction, we design Tandem Attention Module(TAM) and Parallel Attention Module(PAM). TAM is designed to sequentially filter some features from channels and spaces for adaptive feature refinement. PAM combines context information and uses high-level features to guide low-level features to select more distinguishable features. Finally, based on the binary cross entropy loss function, we add an evaluation index to reduce the error caused by determining the optimization direction only through cross entropy. On the Inria Aerial Image Labeling Dataset, FMAM-Net achieves mean IoU of 85.34%, which is 5.58% higher than AMUNet and 3.77% higher than our baseline(U-Net ResNet-34). On the WHU Dataset, IoU reached the maximum value of 91.06% on FMAM-Net, 1.67% higher than SARB-UNet and 0.2% higher than MAP-Net. The visualization results show that FMAM-Net improves the fuzzy boundary of building segmentation and reduces the inter-class indistinction.

关键词： Remote sensing image building segmentation attention mechanism feature refinement encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

R-GRU: Regularized gated recurrent unit for handwritten mathematical expression recognition

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2022年第22期81卷 31405-31419页

作者： Pal, Aniket Singh, Krishna Pratap Indian Inst Informat Technol Dept Informat Technol Machine Learning & Optimizat Lab Allahabad Prayagraj India

Handwritten mathematical expression recognition(HMER) plays a wide variety of roles in different domains like online teaching, scientific research, etc. Due to its two-dimensional non-linear structure, it is a challenging problem. In this work, we proposed a novel architecture using regularization(dropout), attention & gated recurrent unit(R-GRU). R-GRU is used as the central decoding unit that takes the intermediate representation produced by the encoder as an input and generates corresponding LaTeX sequences. Simulation on CROHME 2014 and 2016 datasets achieve results comparable to the latest state-of-the-art.

关键词： Handwritten Mathematical Expression Recognition encoder-decoder Gated recurrent unit Dropout 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：