检索结果-内蒙古大学图书馆

6th international conference on Recent Trends in image processing and pattern recognition, RTIP2R 2023

作者： Pamod, Dilshan Charles, Joseph Hewarathna, Ashen Iranga Vigneshwaran, Palanisamy Lekamge, Sugeeswari Thuseethan, Selvarajah Department of Computing and Information Systems Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka Erlangen Germany Department of Software Engineering Sabaragamuwa University of Sri Lanka Belihuloya Sri Lanka School of Information Technology Deakin University Geelong Australia

ISBN: (纸本)9783031530814

The ability to recognise and interpret emotional expressions is crucial since emotions play a significant role in our daily lives. Emotions are multifaceted phenomena that affect our behavior, perception, and cognition. As a result, numerous machine-learning and deep-learning algorithms for emotion analysis have been studied in previous works. Finding emotion in an obscured face, such as one covered by a scarf or hidden in shadow, is considerably harder than in a complete face, though. This study explores the effectiveness of deep learning models in occluded facial emotion analysis through a transfer learning approach. The performance of two individual pre-trained models, MobileNetV2 and EfficientNetB3, is compared alongside a hybrid model that combines both approaches. This comparison is conducted using the FER-2013 dataset. The dataset consists of 35,887 images and categorizes emotions into seven emotional categories. The results indicate that the hybrid model attained the highest accuracy, with a score of 93.04% for faces occluded at the top and 92.63% for faces occluded at the bottom. Additionally, the study suggests that top-occluded faces displayed more pronounced emotional expressions in comparison to bottom-occluded faces. Overall, these findings imply that hybrid architecture, which was developed as a state-of-the-art model in the study, proves to be effective for analyzing emotions in facial expressions that are partially obscured. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Empirical Analysis of Benchmark Generation for the Verification of Neural Network image Classifiers 1st

Empirical Analysis of Benchmark Generation for the Verifi...

引用

1st international conference on Bridging the Gap between AI and Reality, AISoLA 2023

作者： Manzanas Lopez, Diego Johnson, Taylor T. Vanderbilt University NashvilleTN United States

ISBN: (纸本)9783031460012

Deep learning success in a wide range of applications, such as image recognition and natural language processing, has led to the increasing usage of this technology in many domains, including safety-critical applications such as autonomous cars and medicine. The usage of the models, e.g., neural networks, in safety critical applications demands a thorough evaluation from a component and system level perspective. In these domains, formal methods have the ability to guarantee the correct operation of these components. Despite great efforts in the formal verification of neural networks in the past decade, several challenges remain. One of these challenges is the development of neural networks for easier verification. In this work, we present an empirical analysis, presented as a Latin Hypercube experiment design, in which we evaluate how regularization and initialization methods across different random seeds on two datasets affect the verification analysis of a reachability analysis technique for the verification of neural networks. We show that there are certain training routines that simplify the formal verification task. Lastly, a discussion on how these training approaches impact the robustness verification and reachability computation of the method utilized is included. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Deep learning Formal Verification Medical Imaging Reachability Analysis

来源：评论

学校读者我要写书评

暂无评论

Disease Detection Based on Nail Color Analysis Using image processing 1

Disease Detection Based on Nail Color Analysis Using Image P...

引用

1st international conference on Computational Science and Technology, ICCst 2022

作者： Kumar, D.Sathish Sherly, J.Deepthi Priyadharshini, S. Indira Sri Sai Ram Engineering College Department of Cse Tamilnadu Chennai India

ISBN: (纸本)9781665476553

There have been countless advancements in medical research, and a wide variety of diagnostic procedures for human body problems have been developed. Pathological testing, on the other hand, is time-consuming, painful, and difficult to employ in illness detection. Currently, the patient's nail is scanned and the image is divided. Experiment results demonstrate a considerable impact. The major drawback of this technique is that variations in blood flow may be detected by looking at the color of the image rather than segmentation. Dermatology studies have shown that nail color changes are a good indicator of a variety of health issues, and our suggested method is to use nail color analysis to look for signs of illness. An abnormality in one's health is indicated by a change in the color of one's nails. Heavy equipment and scanning processes are not necessary to take pictures of human nails for use in detection. In contrast to human sight, computer image processing can distinguish even the tiniest differences in color and hence provides more exact and accurate findings. machine learning algorithms such as the CNN (Convolution Neural Network) algorithm are used in the suggested system, which takes an image and assigns value to distinct items in the picture. Using Picture processing, an image is digitally transformed and a few procedures are carried out to extract valuable information. To identify nail traits, image analysis and digital image processing techniques such picture acquisition, image pre-processing, segmentation, and feature extraction are utilized.. Based on variations in nail color, this web tool allows the doctor to detect diseases at an early stage. © 2022 IEEE.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Exploring the Potential of Convolution Neural Network Based image Classification

Exploring the Potential of Convolution Neural Network Based ...

引用

Artificial Intelligence in Education and Industry 4.0 (IDICAIEI), DMIHER international conference on

作者： Ritika Sharma Saimul Bashir Vinay Nath Tiwary Sanjeev Kumar Computer Science and Engineering Chandigarh University Punjab India

This research examines the utilization of Convolutional Neural Networks (CNNs) in image classification, shedding light on recent progress in machine learning algorithms. Recognized as a unique component of deep learning, CNNs have earned significant acclaim in a myriad of visual tasks, attributed to their capacity for hierarchical pattern recognition and extraction of stable features. In this investigation, an enhanced CNN model is introduced, which employs cutting-edge preprocessing and augmentation strategies to elevate image readability, thereby streamlining the learning mechanism. key machine learning principles were employed, equipping the CNN model to identify and distinguish intricate structures within images with amplified precision. The model's efficiency is assessed using multiple benchmark datasets, achieving optimistic outcomes concerning precision, recall, and F1-score. The paper also touches upon inherent obstacles such as overfitting and computational expenses linked with CNNs, suggesting remedies to boost scalability and efficiency. The observations suggest that with meticulous planning and execution, CNNs can serve as a potent instrument for automated image classification, paving the way for a wide array of applications from medical imaging to self-driving vehicles. This research augments the expanding literature in machine learning, offering perspectives that could inform the design of more effective and efficient image classification systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Deep Pixel Distribution learning for Concrete Crack Detection 25

Multi-Scale Deep Pixel Distribution Learning for Concrete Cr...

引用

25th international conference on pattern recognition (ICPR)

作者： Wu, Xuanyi Ma, Jianfei Sun, Yu Zhao, Chenqiu Basu, Anup Univ Alberta Edmonton AB Canada

ISBN: (纸本)9781728188089

A number of methods including image processing technologies (IPTs) and deep learning methods, have been used to detect defects in civilian infrastructure. These methods have been introduced to extract features representing cracks in concrete surfaces. Inspired by recent advances of a pixel distribution learning method in background subtraction, we propose a novel multi-scale deep learning method (MS-DPDL) for concrete crack detection. The designed CNN network is trained on the dataset CRACK500 [1], [2] and tested on it for concrete segmentation. To show good transferability of our proposed model, it is later tested on the dataset Concrete Crack images for Classification [3]. Several existing deep learning methods are used to compare the performance of the proposed MS-DPDL method. Results show that our method has good performance and can effectively find concrete cracks in practical situations.

关键词： image processing technologies (IPTs) deep learning pixel distribution learning CNN concrete crack detection

来源：评论

学校读者我要写书评

暂无评论

image Caption Generation in Kannada using Deep learning Frameworks 1

Image Caption Generation in Kannada using Deep Learning Fram...

引用

1st IEEE international conference on Advances in Electronics, Communication, Computing and Intelligent Information Systems, ICAECIS 2023

作者： Chethas, K. Ankita, V. Apoorva, B.S. Sushma, H. Jayashree, R. Pes University Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798350348057

The official language of the state of Karnataka and one of the most frequently spoken languages in India, Kannada is spoken by more than 57 million people worldwide. In this modern era, one of the most important tasks in the field of Natural Language processing in the present times is the development of image captions. In order to provide a textual description of an image intended to highlight its key qualities, image Caption generation has long been a study of significant interest to academics in the fields of Artificial Intelligence and machine learning. While a great deal of research is being conducted in English, resources in regional languages like Kannada, are severely lacking. Given that 72% of the Karnataka population are Kannada speakers, image Caption generation in Kannada is the realization of human-computer interaction for native language speakers, enabling the development of regional language virtual assistants which is a breakthrough in this line of work. Through our research, we present a model that generates semantic, syntactic and structured captions in Kannada. © 2023 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Underwater image Enhancement Using Pre-trained Transformer 21st

Underwater Image Enhancement Using Pre-trained Transformer

引用

21st international conference on image Analysis and processing (ICIAP)

作者： Boudiaf, Abderrahmene Guo, Yuhang Ghimire, Adarsh Werghi, Naoufel De Masi, Giulia Javed, Sajid Dias, Jorge Khalifa Univ Abu Dhabi U Arab Emirates Technol Innovat Inst ARRC Abu Dhabi U Arab Emirates

ISBN: (纸本)9783031064333;9783031064326

The goal of this work is to apply a denoising image transformer to remove the distortion from underwater images and compare it with other similar approaches. Automatic restoration of underwater images plays an important role since it allows to increase the quality of the images, without the need for more expensive equipment. This is a critical example of the important role of the machine learning algorithms to support marine exploration and monitoring, reducing the need for human intervention like the manual processing of the images, thus saving time, effort, and cost. This paper is the first application of the image transformer-based approach called "Pre-Trained image processing Transformer" to underwater images. This approach is tested on the UFO-120 dataset, containing 1500 images with the corresponding clean images.

关键词： Vision transformer Underwater imaging image enhancement

来源：评论

学校读者我要写书评

暂无评论

Metric learning-Based Unsupervised Domain Adaptation for 3D Skeleton Hand Activities Categorization 21st

Metric Learning-Based Unsupervised Domain Adaptation for 3D ...

引用

21st international conference on image Analysis and processing (ICIAP)

作者： Boutaleb, Yasser Soladie, Catherine Duong, Nam-duong Kacete, Amine Royan, Jerome Seguier, Renaud IRT B Com 1219 Ave Champs Blancs F-35510 Cesson Sevigne France IETR Cent Supelec Ave Boulaie F-35510 Cesson Sevigne France

ISBN: (纸本)9783031064333;9783031064326

First-person hand activity recognition plays a significant role in the computer vision field with various applications. Thanks to recent advances in depth sensors, several 3D skeleton-based hand activity recognition methods using supervised Deep learning (DL) have been proposed, proven effective when a large amount of labeled data is available. However, the annotation of such data remains difficult and costly, which motivates the use of unsupervised methods. We propose in this paper a new approach based on unsupervised domain adaptation (UDA) for 3D skeleton hand activity clustering. It aims at exploiting the knowledge-driven from labeled samples of the source domain to categorize the unlabeled ones of the target domain. To this end, we introduce a novel metric learning-based loss function to learn a highly discriminative representation while preserving a good activity recognition accuracy on the source domain. The learned representation is used as a low-level manifold to cluster unlabeled samples. In addition, to ensure the best clustering results, we proposed a statistical and consensus-clustering-based strategy. The proposed approach is experimented on the real-world FPHA data set.

关键词： Unsupervised domain adaptation Metric learning Hand activities clustering

来源：评论

学校读者我要写书评

暂无评论

Gesture recognition of the Kazakh alphabet based on machine and deep learning models

引用

Procedia Computer Science 2024年 241卷 458-463页

作者： Samat Mukhanov Raissa Uskenbayeva Abd. A. Rakhim Akbota Akim Symbat Mamanova International Information Technology University Manas St. 34/1 Almaty 050043 Kazakhstan Satpayev University Satpayev St. 22 Almaty 050000 Kazakhstan Universiti Tenaga Nasional Kajang 50000 Malaysia

Currently, a growing body of research focuses on addressing problems using computer vision libraries and artificial intelligence tools. The predominant approaches involve employing machine and deep learning models of artificial neural networks to recognize gestures in the Kazakh Sign Alphabet (KSA) via supervised and deep learning techniques for sequential data processing. pattern recognition in this context involves identifying an object within an image, where the object can be abstract and vary in shape. We have chosen to investigate the field of gesture recognition, specifically. For recognizing Kazakh Sign Language (KSL), the initial step involves mastering the KSA. Training a neural network to recognize KSL necessitates the collection of datasets in the form of images depicting hand gestures. In this research, prominent hand gesture recognition models such as the Convolutional Neural Network (CNN), Long Short-Term Memory (LstM), and Support Vector machine (SVM) were analyzed. These models differ in their methodologies, processing times, and training data requirements. A significant aspect of this study is the application of unsupervised and supervised learning techniques including CNN, LstM, and SVM. The experiments yielded diverse results when training neural networks for recognizing gestures in Kazakh sign language based on the dactyl alphabet. This article provides a comprehensive overview of each method, their specific purposes, and their effectiveness in terms of performance and training. Numerous experimental outcomes were documented in a table, showcasing the accuracy of recognizing each gesture. Additionally, specific hand gestures were tested in front of a camera to identify the gesture and display the result on the screen. A notable feature was the use of mathematical formulas and functions to elucidate the operating principles of the machine learning methods, as well as the logical structure and design of the LstM model.

关键词： Hand gesture recogtion Kazakh Sign Language (KSL) neural netwrok SVM LstM

来源：评论

学校读者我要写书评

暂无评论

Hand image Based Skin Disease Identification Using machine learning and Deep learning Algorithms 1

Hand Image Based Skin Disease Identification Using Machine L...

引用

1st international conference on Technologies for Smart Green Connected Society 2021, ICTSGS 2021

作者： Kalaiyarivu, M. Nalini, N.J. Department of Computer and Information Science Annamalai University India

ISBN: (纸本)9781607685395

The skin disease is most commonly affected to all peoples. It is a big challenging process in skin image identification because of the combined texture shape with the color variations. This paper presented to identify a skin using texture features(Local Binary Patteren and Gray Level Cooccurence Matrix) and Color features with various learning models. The proposed work technique for identification of skin disease has two stages: Either Extracted feature values and identification. During the first stage, Texture features and Color feature vector values are taken from the images in a dataset of hand skin database. In during the second stage, the extracted feature vectors are trained by the various kinds of machine learning models. This hand skin image dataset also applied into Deep learning model for identification. The performance measures of the system is evaluated using accuracy measure. The Convolutional Neural Network(CNN) model gives accuracy of 87.5% when compared to machine learning models. © The Electrochemical Society

关键词： Color Features Deep learning machine learning Pre-processing Segmentation Texture Features

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：