检索结果-内蒙古大学图书馆

International Joint Conference on Computer Science and Software Engineering (JCSSE)

作者： Possathon Teerakantapirut Rungrat Viratikul Charnchai Pluempittiwiriyawej Department of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Chulalongkorn University Bangkok Thailand Department of Electrical Engineering Telecommunication Networking and Systems Chulalongkorn University Bangkok Thailand

ISBN: (数字)9798350381764

ISBN: (纸本)9798350381771

This study introduces an advanced methodology to overcome the intrinsic challenges associated with Light Detection and Ranging (LiDAR) technology integrated into Apple's devices, specifically focusing on optimizing depth map accuracy through multiview angle analysis. Despite the extensive utility of LiDAR in generating detailed three-dimensional models for various applications, from autonomous vehicle navigation to environmental conservation, the efficiency of the technology is often compromised by difficulties in accurately rendering complex geometries from multiple perspectives and by its substantial demands on computational resources. To address these limitations, depth map data are systematically collected and evaluated from four distinct viewing angles (0, 90, 180, and 270 degrees) against established ground truth benchmarks. With a threshold of Intersection over Union (IoU) set at 0.8, it is inferred that the predicted shape exhibits over 80% similarity to the actual shape. When only one view is utilized, an IoU of 0.95 is achieved, alongside a size reduction of 73.07%, a Perimeter Accuracy reduction of 1.56%, and an increase in Root Mean Square Error (RMSE) of 29.37%. This comprehensive analysis highlights the critical need for precise geometric representation and marks a significant step toward enhancing the fidelity of LiDAR measurements. The results of this research indicate a substantial improvement in the accuracy of three-dimensional representations, suggesting a potential shift in the utilization of LiDAR technology across various scientific and infrastructural domains. By proposing a novel framework that optimizes data precision and processing efficiency, this paper contributes to the ongoing discourse on improving 3D modeling and analysis techniques.

关键词： Geometry Solid modeling Laser radar Accuracy Three-dimensional displays Shape Shape measurement

来源：评论

学校读者我要写书评

暂无评论

Compressed Image Super Resolution using Convolutional Neural Network

Compressed Image Super Resolution using Convolutional Neural...

引用

IEEE Region 10 International Conference TENCON

作者： Ei Ei Tun Aktanin Konkitkriengkrai Watchara Ruangsang Supavadee Aramvith Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand Department of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Faculty of Engineering Chulalongkorn University Bangkok Thailand

Image compression is a topic of significant interest as it reduces file sizes in stored data. In this paper, we propose a model that achieves multiple levels of compression, thereby minimizing the storage space required for images, which typically consume substantial amounts of data due to their size and resolution. We combine an image downscaling and upscaling model with an image compression model. By leveraging convolutional techniques to identify image features, we can effectively reduce the size of the image through downscaling and subsequently upscaling it. Additionally, we employ entropy image compression and arithmetic encoding to compress and reconstruct the image while preserving its lossless data. Through experimentation with the Kodak dataset, we observed that our proposed model achieved a compression rate of 96.92%, significantly reducing the data needed for file storage. Moreover, our reconstructed images attained a standardized measure with a signal-to-noise ratio of 33.10 dB and a structural similarity of 0.9219. Notably, the perceptual quality of the images, including intricate details, remained intact to the human eye.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multiple Face Detection and Recognition for System-an-Chip FPGAs

Multiple Face Detection and Recognition for System-an-Chip F...

引用

International Joint Conference on Computer Science and Software Engineering (JCSSE)

作者： Siraphop Santiwiwat Amir Hajian Watchara Ruangsang Supavadee Aramvith Department of Electrical Engineering Multimedia Data Analytic and Processing Research Unit Faculty of Engineering Chulalongkorn University Bangkok Thailand

ISBN: (数字)9798350381764

ISBN: (纸本)9798350381771

Recognizing multiple faces within a single frame or image presents a significant challenge in facial recognition tasks. This challenge demands robust algorithms capable of handling variations in facial position, unstable brightness, and the presence of accessories such as glasses or hats. However, in current surveillance applications, multiple face recognition is vital for maintaining various security measures. In such applications, there is a need for processors that are compact and deployable in diverse environments. FPGA is popular due to its compact size, cost-effectiveness, and high performance. This paper presents the implementation of multiple face detection and recognition on an SoC FPGA utilizing the Xilinx Kria KV260 platform. We chose the MTCNN algorithm for multiple face detection and the ResNet-20 model for multiple face recognition developed by Vitis AI, and we fine-tuned this model using the CUFace dataset. Our results demonstrate the robust capability to detect and recognize faces of up to 8 individuals within a single frame in just 0.11 seconds, with a recognition rate of 93.77%.

关键词： Program processors Image recognition Face recognition Surveillance Face detection Security Task analysis

来源：评论

学校读者我要写书评

暂无评论

YOLO for Small Objects in Aerial Imagery: A Performance Evaluation

YOLO for Small Objects in Aerial Imagery: A Performance Eval...

引用

International Joint Conference on Computer Science and Software Engineering (JCSSE)

作者： Wimonthip Saenprasert Ei Ei Tun Amir Hajian Watchara Ruangsang Supavadee Aramvith Department of Electrical Engineering Multimedia Data Analytic and Processing Research Unit Faculty of Engineering Chulalongkorn University Bangkok Thailand

ISBN: (数字)9798350381764

ISBN: (纸本)9798350381771

Unmanned aerial vehicles (UAVs) offer significant advantages in accessing remote or high-risk locations. Yet, they also present challenges concerning privacy and safety. Effective air traffic management is crucial for ensuring compliance with security regulations. The primary objective is to monitor the flight trajectories of target objects in the airspace, necessitating efficient object detection, especially for distant objects, to facilitate surveillance and preparedness for addressing potential threats. However, the current performance of existing models in detecting small objects is inadequate for real-time surveillance applications. This study advocates prioritizing object detection over precise prediction of object bounding box sizes. We propose a novel approach where the efficacy of object detection is measured by the ability to identify small objects. This research serves as an additional variable for evaluating and enhancing models designed for small object detection. We selected several versions of the YOLO model and assessed them on the COCO val 2017 dataset, counting the number of accurately predicted objects. Based on the most popular drone size of 100–400 pixels, YOLOv5-x is best suited for the future development of smaller objects. However, because the model is large, we can choose the YOLOv5-s, which is the smaller version. YOLOv5-s detects small objects in the same way that the newest version of the YOLO family, the YOLOv9-e. By measuring this, we may establish the model's capabilities and select a model appropriate for the target object's size.

关键词： YOLO Accuracy Computational modeling Surveillance Prototypes Real-time systems Trajectory

来源：评论

学校读者我要写书评

暂无评论

Enhancing Visual Fidelity in Unpaired Night-to-Day Image Translation through a Perceptual Quality Focused Training Objective

Enhancing Visual Fidelity in Unpaired Night-to-Day Image Tra...

引用

Artificial Intelligence (SLAAI-ICAI), SLAAI International Conference on

作者： H.K.I.S. Lakmal Maheshi B. Dissanayake Supavadee Aramvith Dept. of Electrical and Electronic Engineering Faculty of Engineering University of Peradeniya Peradeniya Sri Lanka Dept. of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Chulalongkorn University Bangkok Thailand

Nighttime driving poses visibility challenges, but image translation methods can help by transforming night images into day-like scenes. The Cycle-GAN is a versatile unpaired image translation model which can easily be adept to night-to-day image translation tasks. However, it generates unnatural and unrealistic outcomes in these cases. This study is focused on addressing this with a novel training strategy for the Cycle-GAN, employing a tailored training objective that incorporates perceptual quality optimization. This training objective aims to boost the naturalness and perceptual quality of the generated images by preserving high-level image features. The optimization process involves minimizing Euclidean distances between synthesized and target image feature maps, which are derived from the pre-trained VGG19 network. Experimental findings attest to the effectiveness of this method, revealing noteworthy improvements of 8% in Inception Score, 2% in NIQE Score, and 13% in BRISQUE Score.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Combined Channel and Spatial Attention-based Stereo Endoscopic Image Super-Resolution

arXiv

引用

arXiv 2023年

作者： Hayat, Mansoor Armvith, Supavadee Achakulvisut, Titipat Department of Electrical Engineering Chulalongkorn University Bangkok Thailand Multimedia Data Analytics and Processing Research Unit Department of Electrical Engineering Chulalongkorn University Bangkok Thailand Department of Biomedical Engineering Mahidol University Bangkok Thailand

Stereo Imaging technology integration into medical diagnostics and surgeries brings a great revolution in the field of medical sciences. Now, surgeons and physicians have better insight into the anatomy of patients' organs. Like other technologies, stereo cameras have limitations, e.g., low resolution (LR) and blurry output images. Currently, most of the proposed techniques for super-resolution focus on developing complex blocks and complicated loss functions, which cause high system complexity. We proposed a combined channel and spatial attention block to extract features incorporated with a specific but very strong parallax attention module (PAM) for endoscopic image super-resolution. The proposed model is trained using the da Vinci dataset on scales 2 and 4. Our proposed model has improved PSNR up to 2.12 dB for scale 2 and 1.29 dB for scale 4, while SSIM is improved by 0.03 for scale 2 and 0.0008 for scale 4. By incorporating this method, diagnosis and treatment for endoscopic images can be more accurate and effective. Copyright © 2023, The Authors. All rights reserved.

关键词： Geometrical optics

来源：评论

学校读者我要写书评

暂无评论

Combined Channel and Spatial Attention-Based Stereo Endoscopic Image Super-Resolution

Combined Channel and Spatial Attention-Based Stereo Endoscop...

引用

IEEE Region 10 International Conference TENCON

作者： Mansoor Hayat Supavadee Armvith Titipat Achakulvisut Department of Electrical Engineering Chulalongkorn University Bangkok Thailand Department of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Chulalongkorn University Bangkok Thailand Department of Biomedical Engineering Mahidol University Bangkok Thailand

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Based Classification of Mental Health State Using the DASS-21 Profile 16

Machine Learning-Based Classification of Mental Health State...

引用

16th Biomedical Engineering International Conference, BMEiCON 2024

作者： Duangchaemkarn, Khanita Khammarew, Patiphan Aramvith, Supavadee School of Pharmaceutical Sciences University of Phayao Department of Pharmacy Practice Phayao56000 Thailand School of Pharmaceutical Sciences University of Phayao Pharmacy Research Extern Under the Doctor of Pharmacy Program Phayao56000 Thailand Chulalongkorn University Multimedia Data Analytic and Processing Research Unit Department of Electrical Engineering Faculty of Engineering Bangkok10330 Thailand

ISBN: (纸本)9798331505431

This study aimed to determine the most effective machine learning model for classifying emotional states of depression, anxiety, and stress based on the DASS-21 profile. Six machine learning models were developed and tested using 2,602 records from the University of Phayao Counselling Center, collected between 23 July and 16 August 2020. The results demonstrate that the deep learning model provided the highest accuracy and precision, exceeding 90% for all three emotional categories, with an AUC greater than 99%. These findings introduce an innovative approach for classifying emotional states of depression, anxiety, and stress using the DASS-21 profile as a benchmark. Incorporating additional features into the model could further enhance its utility in clinical decision-making, improving the accuracy of mental health screening and diagnosis. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced Cross-Modality MRI Segmentation Using Dilated Convolutions and Multi-Scale Gradient Map

Enhanced Cross-Modality MRI Segmentation Using Dilated Convo...

引用

International Computer Science and Engineering Conference (ICSEC)

作者： Ghulam Murtaza Charnchai Pluempitiwiriyawej Somkiat Wangsiripitak Mohammad Jawad Fareed Mudassar Khalid Department of Electrical Engineering Chulalongkorn University Bangkok Thailand Department of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Chulalongkorn University Bangkok Thailand Machine Intelligence and Vision Laboratory School of Information Technology King Mongkut’s Insittute of Technology Ladkrabang Thailand

ISBN: (数字)9798350366860

ISBN: (纸本)9798350366877

In recent years, convolutional neural networks have significantly advanced image segmentation, particularly for brain images, where important edge features are automatically found. However, accurate segmentation of tumors in a brain remains a challenge across different magnetic resonance modalities, like T1, T2, T1ce, and FLAIR. Using a simple gradient map as an input to the neural networks is not effective due to variations in cross-modality image characteristics. To address this issue, we introduced multi-scale gradient maps that incorporate Holistically Nested Edge Detection (HED) and dilated convolutions into the UNet model. The HED model captures detailed gradient information, enhancing structural feature identification across modalities, while dilated convolutions expand the UNet receptive field for better contextual understanding without increasing parameters. Our method was trained and evaluated on the BraTS2018 dataset. The experimental results demonstrate significant improvements in segmentation accuracy and robustness. Specifically, our method achieved a Dice Similarity Coefficient (DSC) of 0.6902 for T2 to T1ce, 0.6858 for T2 to T1, 0.4329 for FLAIR to T1, and 0.6004 for FLAIR to T1ce, outperforming previous state-of-the-art methods. This demonstrates the effectiveness of our approach in enhancing segmentation performance across different MR image modalities.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Estimation of Eucalyptus DBH from UAV-LiDAR data Utilizing Advanced Point Cloud processing Techniques

Estimation of Eucalyptus DBH from UAV-LiDAR Data Utilizing A...

引用

International Conference on Business and Industrial research (ICBIR)

作者： Peerapong Dangrungroj Monton Udompitaksook Natapong Intarasuk Charnchai Pluempitiwiriyawej Teerapol Silawan Department of Electrical Engineering Communications and Information Engineering (CIE) Laboratory Chulalongkorn University Bangkok Thailand Sky Visual Imaging Venture Co. Ltd Bangkok Thailand Department of Electrical Engineering Multimedia Data Analytics and Processing Research Unit Chulalongkorn University Bangkok Thailand

ISBN: (数字)9798350383027

ISBN: (纸本)9798350383034

The Diameter at Breast Height (DBH) measurements are essential for forest management and carbon absorption estimation in environmental challenges. Traditional DBH measurements require more precision and efficiency, especially in inaccessible forests. Existing methods, including manual measurement or remote sensing, struggle with precision in low-intensity point cloud areas. Our technique solves these limitations by analyzing the entire tree stem by the process, including convex hull construction and advanced noise reduction using real-world datasets. It has reduced the Root Mean Squared Error (RMSE) to 0.020, or 92.25% improvement, and the bias has decreased to -0.008. These enhancements highlight the accuracy and reliability of our DBH estimation technique.

关键词： Point cloud compression Adaptation models Convex hulls Accuracy Noise reduction Estimation Forestry Vegetation Breast Remote sensing

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：