检索结果-内蒙古大学图书馆

20th International Symposium on Medical Information processing and Analysis

作者： Rodriguez Villamizar, Julian Orlando Rueda Marino, Daniel Felipe Romo-Bucheli, David Univ Ind Santander Escuela Ingn Sistemas & Informat Bucaramanga Colombia

ISBN: (纸本)9798331528669;9798331528652

In histological image analysis, an expert pathologist might spend a considerable amount of time assessing the anatomical features of the tissues that contain cancerous cells to determine their malignancy. It is common practice to first identify clinically relevant regions at a lower magnification and subsequently analyze the tissue at higher magnification levels. Machine learning techniques offer the potential to streamline these tasks by directly classifying the tissue. However, the interpretation of the results of these models is still an open problem. Our study aims to identify if attended regions in recurrent visual models corresponds to clinical regions of interest defined by experts. Similarly, it is necessary to verify that the model is capable of adequately solving the classification problem to correctly identify images with healthy or cancerous tissues. In our work, we used a recurrent neural network-based image classifier capable of sequentially focus its attention on the most representative areas of histological images to classify histological slides. Even when dealing with a large volume of high-magnification image data, the model can identify relevant regions within the image. We hypothesize that the attended locations should be highly associated with clinically relevant areas in the histological image. We evaluated this hypothesis by measuring the spatial overlapping of the attended locations via probability density divergence metrics, such as Kullback Leibler Divergence, Jensen Shannon Divergence, and Mutual information. The obtained results do not allow us to assert that in general the model's attended regions correspond to clinical regions. However, it was found that for those cases in which the model was able to perform better, in terms of the mentioned metrics, the annotations were more precise and included a more defined area (covering on average 10% of the tissue). Meanwhile, those cases in which a uniform sampling strategy perform better, the a

关键词： deep Recurrent Attention Model Visual Attention Histological images Cancer

来源：评论

学校读者我要写书评

暂无评论

A Cycle-GAN-Based Method for real-World image Super-Resolution 13

A Cycle-GAN-Based Method for Real-World Image Super-Resoluti...

引用

13th IEEE Global Conference on Consumer Electronic, GCCE 2024

作者： Imamura, Ryusei Li, Yinhao Taga, Hiroshi Iwasa, Koki Shichikawa, Ryuichi Suganami, Makoto Nakamoto, Kazuhiro Chen, Yen-Wei Ritsumeikan University Graduate School of Information Science and Engineering Osaka Japan Ritsumeikan University School of Information Science and Engineering Osaka Japan Mitsui E&s Systems Research Inc. Dx Promotion Department Chiba Japan Mitsui E&s Systems Research Inc. Industry Systems Solutions Division Chiba Japan

ISBN: (纸本)9798350355079

In recent years, the development of deep learning-based image super-resolution (SR) technology has made it easier to generate high-resolution, clear images. However, in practical scenarios where cameras cannot be placed close to objects, the captured image quality significantly degrades, leading to very low accuracy in automatic recognition of digits and text. Traditional SR methods still fail to achieve satisfactory results in such situations. Many supervised SR studies use synthetic low-resolution (LR) images created from high-resolution (HR) images for training. However, synthetic LR images do not correspond to real-world degradation. Therefore, to reconstruct degraded images captured in real-world scenarios for high-accuracy automatic character recognition, this study generates low-quality images that simulate real-world conditions to train SR models, thereby restoring and enhancing characters in degraded captured images. Our approach comprises two parts: domain correction and SR model training. First, we use CycleGAN to synthesize LR images that include real-world degradation features. Second, the generated LR images and HR images are paired to train the SR model. To validate the effectiveness of the proposed method, we used YOLOv5 for text recognition before and after SR processing. Experimental results show that our method effectively improves image quality and the accuracy of optical characteristic recognition. © 2024 IEEE.

关键词： image quality

来源：评论

学校读者我要写书评

暂无评论

Research on Yak Behavior Recognition Algorithm Based on deep learning 9

Research on Yak Behavior Recognition Algorithm Based on Deep...

引用

9th International Conference on Intelligent Computing and Signal processing, ICSP 2024

作者： Shen, Yichao Shen, Jianwei Macao Polytechnic University China New Energy and Power System Research Center Ningbo Polytechnic Ningbo China

ISBN: (纸本)9798350376548

Yaks play a crucial role in the survival of pastoral communities and have made significant contributions to the fragile ecosystems in high-altitude areas. Tracking their behaviour can provide important information about their health, welfare, and performance. This paper proposes a method for yak detection using Faster R-CNN and behaviour recognition using SlowFast. The model was trained and evaluated on a custom dataset comprising images and videos of yaks in various environments, including feeding, standing, and walking behaviours. The proposed method is suitable for real-time monitoring and management of yaks in various situations. The proposed method comprises three stages: dataset creation and yak detection using Faster R-CNN, and behaviour resdddsdscognition using SlowFast. During the target detection dataset creation phase, a number of typical yak behaviours, such as feeding, standing, lying, running, and walking, were selected. The dataset was then enriched through data augmentation. During the behavior recognition phase, the SlowFast network was trained on the yak video dataset that was prepared through frame segmentation, annotation, and optimization to achieve higher accuracy. The results indicate that the proposed method achieved an accuracy rate exceeding 67% on eating and standing behaviour. © 2024 IEEE.

关键词： deep learning

来源：评论

学校读者我要写书评

暂无评论

SimICL: A Simple Visual In-context learning Framework for Ultrasound Segmentation 46

SimICL: A Simple Visual In-context Learning Framework for Ul...

引用

46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024

作者： Zhou, Yuyue Felfeliyan, Banafshe Ghosh, Shrimanti Knight, Jessica Alves-Pereira, Fatima Keen, Christopher Kupper, Jessica Hareendranathan, Abhilash Rakkunedeth Jaremko, Jacob L. University of Alberta Department of Radiology and Diagnostic Imaging Edmonton Canada University of Alberta Department of Biomedical Engineering Edmonton Canada

ISBN: (纸本)9798350371499

Conventional deep learning models deal with images one-by-one, requiring costly and time-consuming expert labeling in the field of medical imaging, and domain-specific restriction limits model generalizability. Visual in-context learning (ICL) is a new and exciting area of research in computer vision. Unlike conventional deep learning, ICL emphasizes the model's ability to adapt to new tasks based on given examples quickly. Inspired by MAE-VQGAN, we proposed a new simple visual ICL method called SimICL, combining visual ICL pairing images with masked image modeling (MIM) designed for self-supervised learning. We validated our method on bony structures segmentation in a wrist ultrasound (US) dataset with limited annotations, where the clinical objective was to segment bony structures to help with further fracture detection. We used a test set containing 3822 images from 18 patients for bony region segmentation. SimICL achieved an remarkably high Dice coeffient (DC) of 0.96 and Jaccard Index (IoU) of 0.92, surpassing state-of-the-art segmentation and visual ICL models (a maximum DC 0.86 and IoU 0.76), with SimICL DC and IoU increasing by at least 0.10 and 0.16 respectively. This remarkably high agreement with limited manual annotations indicates SimICL could be used for training AI models even on small US datasets. This could dramatically decrease the human expert time required for image labeling compared to conventional approaches, and enhance the real-world use of AI assistance in US image analysis. © 2024 IEEE.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

COCO-TEACH: A CONTRASTIVE CO-TEACHING NETWORK FOR INCREMENTAL 3D OBJECT DETECTION 30

COCO-TEACH: A CONTRASTIVE CO-TEACHING NETWORK FOR INCREMENTA...

引用

30th IEEE International Conference on image processing (ICIP)

作者： Cheng, Zhongyao Chen, Cen Zhao, Ziyuan Qian, Peisheng Li, Xiaoli Yang, Xulei ASTAR Inst Infocomm Res I2R Singapore Singapore

ISBN: (纸本)9781728198354

deep learning (DL) models for 3D object detection from point clouds have shown remarkable progress in various autonomous perception scenarios. However, the issue of catastrophic forgetting seriously hinders the deployment of these models in real-world applications where new classes are encountered over time. In order to address this issue, we present the Contrastive Co-Teaching Network (COCO-TEACH) framework for class-incremental 3D object detection. Our proposed framework consists of two teacher networks: a primary teacher network that detects old class objects in new data and provides them with pseudo-labels and an auxiliary teacher network that leverages the unlabelled objects in new data. The two teacher models transfer their learned knowledge to the target student model through a class-aware consistency loss. To enhance this transfer, a supervised contrastive loss is further incorporated into the loss function. We evaluate the performance of our proposed method against baseline methods through extensive experiments on two benchmark datasets. The results show that our proposed framework achieves state-of-the-art performance on incremental 3D object detection.

关键词： deep learning Incremental learning Knowledge distillation Point cloud data 3D object detection

来源：评论

学校读者我要写书评

暂无评论

Enhancing Energy Efficiency and Security in Wireless Sensor Networks: A Survey and Proposed Methodology Integrating Prediction Algorithms and deep Neural Networks 6

Enhancing Energy Efficiency and Security in Wireless Sensor ...

引用

6th International Conference on Cybernetics, Cognition and Machine learning Applications, ICCCMLA 2024

作者： Shafiuddin, Shaik Krishna, Konda Hari Mohan Babu University School of Computing Department of Computer Science and Engineering Andhra Pradesh Tirupati India

ISBN: (纸本)9798331505790

This research study provides a complete assessment of energy- and trust-aware techniques in IoT-WSNs, emphasizing the significant problems and limits of current methodologies. Traditional methodologies frequently face challenges such as excessive processing overhead, difficulty reaching real-time optimal solutions, and increased delay, especially in large-scale or resource-constrained contexts. To solve these issues, the paper suggests a new approach that combines powerful prediction algorithms, deep neural networks, and strong security frameworks. The process entails initializing exact inputs and simulation settings, modeling advanced prediction algorithms, developing deep neural networks, and constructing resilient sensor nodes, all while implementing extensive security measures. The suggested approach seeks to considerably improve energy efficiency, network lifetime, and security by focusing on mobile sink movement prediction and network operation optimization. This study not only covers existing methodologies, but also shows the promise of the proposed methodology through extensive analysis and preliminary results, demonstrating its ability to overcome the limits of present WSN deployments. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Sherlock-A flexible, low-resource tool for processing camera-trapping images

引用

METHODS IN ECOLOGY AND EVOLUTION 2024年第1期15卷 91-102页

作者： Penn, Matthew J. Miles, Verity Astley, Kelly L. Ham, Cally Woodroffe, Rosie Rowcliffe, Marcus Donnelly, Christl A. Univ Oxford Dept Stat Oxford England Inst Zool London England Univ Oxford Pandem Sci Inst Oxford England

The use of camera traps to study wildlife has increased markedly in the last two decades. Camera surveys typically produce large data sets which require processing to isolate images containing the species of interest. This is time consuming and costly, particularly if there are many empty images that can result from false triggers. Computer vision technology can assist with data processing, but existing artificial intelligence algorithms are limited by the requirement of a training data set, which itself can be challenging to acquire. Furthermore, deep-learning methods often require powerful hardware and proficient coding *** present Sherlock, a novel algorithm that can reduce the time required to process camera trap data by removing a large number of unwanted images. The code is adaptable, simple to use and requires minimal processing *** tested Sherlock on 240,596 camera trap images collected from 46 cameras placed in a range of habitats on farms in Cornwall, United Kingdom, and set the parameters to find European badgers (Meles meles). The algorithm correctly classified 91.9% of badger images and removed 49.3% of the unwanted 'empty' images. When testing model parameters, we found that faster processing times were achieved by reducing both the number of sampled pixels and 'bouncing' attempts (the number of paths explored to identify a disturbance), with minimal implications for model sensitivity and specificity. When Sherlock was tested on two sites which contained no livestock in their images, its performance greatly improved and it removed 92.3% of the empty *** further refinements may improve its performance, Sherlock is currently an accessible, simple and useful tool for processing camera trap data.

关键词： camera-trapping image classification

来源：评论

学校读者我要写书评

暂无评论

Vision Based Runway Identification with marked or unmarked terrain for Automatic Landing Applications of UAV 8

Vision Based Runway Identification with marked or unmarked t...

引用

8th Conference on Advances in Control and Optimization of Dynamical Systems (ACODS)

作者： Tripathi, Amit Kumar Patel, Vijay V. Padhi, Radhakant Aeronaut Dev Agcy Bangalore Karnataka India Indian Inst Sci Fac Aerosp Engn Dept Bangalore Karnataka India

Vision based runway identification using 'marked or unmarked terrain' image sequences captured from a fixed wing unmanned aerial vehicle through onboard stereovision sensor is presented in this paper. An innovative convolutional neural netwok (CNN) based YOLO-V8 object detection algorithm is used to detect the runway during approach segment of UAV. This deep learning algorithm detects the region of interest in real time and in a computationally efficient manner. The captured unknown road segment or runway image frames are processed and examined for width, length, level and smoothness aspects to qualify as a suitable runway for UAV landings. Also, it is ensured that there are no obstacles, patches or holes on the detected road or runway. Runway start and end threshold lines and regions, touchdown point and runway edge lines are considered as the region of interest. image processing algorithms are applied on the captured runway or road images to detect strong features in the region of interest. Feature detector based image processing algorithm with stereo vision constraint is used to establish the relation between unmanned aerial vehicle's center of gravity and detected runway feature points image processing algorithms like hough line detection, RANSAC, Oriented FAST and Rotated BRIEF (ORB), median filters, morphological methods are applied to extract terrain features. Based on the detected runway orientation and position with respect to UAV position. An automatic landing manoeuvre is performed by UAV autopilot to land the UAV on intended touchdown point on runway computed through detected feature points.

关键词： image processing Runway Identification Feature detection Region of Interest threshold lines touch down point YOLOv8 ORB CNN runway width slant distance

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection in Spacecraft Telemetry: Forecasting vs. Classification 15

Anomaly Detection in Spacecraft Telemetry: Forecasting vs. C...

引用

15th Space Computing Conference

作者： Lakey, Daniel Schlippe, Tim IU Int Univ Appl Sci Berlin Germany

ISBN: (纸本)9798350384543;9798350384536

Anomaly detection in spacecraft telemetry is critical for the success and safety of space missions. Traditional methods often rely on forecasting and threshold techniques to identify anomalies [1]-[5]. This paper presents a comprehensive comparison of traditional forecast-based anomaly detection against two innovative classification methods, including a direct classification and an image classification through Gramian Angular Field (GAF) transforms [6], which have only been analysed in other domains but not for spacecraft anomaly detection. All our investigated systems leverage deep learning architectures and use the popular real SMAP/MSL spacecraft data from [2]. Our findings suggest that direct classification provides a marginal but statistically significant improvement in anomaly detection over traditional methods. However, image classification, while less successful, offers promising directions for future research. The study aims to guide the selection of appropriate anomaly detection techniques for spacecraft telemetry and contribute to the advancement of automated monitoring systems in space missions.

关键词： anomaly detection time series classification image classification

来源：评论

学校读者我要写书评

暂无评论

Optimizing Sentiment Analysis on Twitter: Leveraging Hybrid deep learning Models for Enhanced Efficiency 1

引用

20th International Conference on Distributed Computing and Intelligent Technology (ICDCIT)

作者： Ashok, Gadde Ruthvik, N. Jeyakumar, G. Coimbatore Amrita Vishwa Vidyapeetham Amrita Sch Comp Dept Comp Sci & Engn Coimbatore India

ISBN: (数字)9783031505836

ISBN: (纸本)9783031505829;9783031505836

Sentiment analysis has emerged as a prominent and critical research area, particularly in the realm of social media platforms. Among these platforms, Twitter stands out as a significant channel where users freely express opinions and emotions on diverse topics, making it a goldmine for understanding public sentiment. The study presented in this paper delves into the profound significance of sentiment analysis within the context of Twitter, with a primary focus on uncovering the underlying sentiments and attitudes of users towards various subjects. To achieve it, this study presents a comprehensive analysis of sentiment on Twitter, leveraging a diverse range of advanced deep learning and neural network models, including Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). Moreover, investigates the effectiveness of Hybrid Ensemble Models in enhancing sentiment analysis accuracy and optimized time. The proposed architecture (HCCRNN) puts forward a sophisticated deep learning model for sentiment analysis on Twitter data, achieves great accuracy whilst considering computational efficiency. Standard models such as Multinomial-NB, CNN, RNN, RNN-LSTM, and RNN-CNN, as well as hybrid models such as HCCRNN (2CNN-1LSTM), CATBOOST, and STACKING (RF-GBC), were examined CNN and RNN-CNN had the best accuracy (82%) and F1-score (81%), with appropriate precision and recall rates among the conventional models. RNN-CNN surpassed other models in terms of analysis time, requiring just 22.4 min. For hybrid models, our suggested model, HCCRNN (2CNN-1LSTM), attained high accuracy in 59 s and an accuracy of 82.6%. It exhibits the capability of real-time sentiment analysis with extraordinary precision and efficiency. This comprehensive exploration of sentiment analysis on Twitter enriches the knowledge base of the community and the application of sentiment analysis across diverse domains.

关键词： Sentiment Analysis Twitter Social media Natural Language processing deep learning Hybrid Ensemble Models Text Classification Emotion Analysis Text Mining

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：