检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Yang Pu Qingfeng Wu School of Film Xiamen University Xiamen China Xiamen University Xiamen China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In this paper, we address the problem of reconstructing geometry, material and illumination from multi-view images of a scene captured in an unknown environment. Compared to prevalent NeRF-based methods, which are typically hindered by slow training and time-intensive evaluating, our approach capitalizes on recent advancements in Volume Rendering: point-based 3D Gaussian Splatting (3DGS) techniques. However, the inherent lack of geometric priors in this point-based representation and the material-lighting ambiguity of the radiance field make it intractable for inverse rendering tasks. To overcome this challenge, our method adopt a two-step approach which we call InvGS. First, our method optimize the original parameters of the initial 3DGS point cloud using flattening and alignment regularization to make the Gaussians closer to the object surfaces for reconstructing a more compact geometry. In the second stage, our method jointly optimize lighting and material parameters of 3DGS point cloud through the differentiable rendering process, thereby following the physically-based rendering equation. Benefiting from the fast rasterization of 3DGS and real-time rendering techniques in physically-based rendering, InvGS can reasonably shade each Gaussian under image-based HDR illumination without compromising real-time performance. We demonstrate that our method is comparable to the state of the arts, and even outperform classic and deep learning-based approaches.

关键词： Point cloud compression Geometry Training Three-dimensional displays Lighting Rendering (computer graphics) real-time systems Vectors image reconstruction Surface treatment

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of palpebral conjunctiva using predefined deep neural architectures for anemia detection

引用

Procedia Computer Science 2023年 218卷 328-337页

作者： Sabrina Dhalla Junaid Maqbool Tanvir Singh Mann Aastha Gupta Ajay Mittal Preeti Aggarwal Krishan Saluja Munish Kumar Shiv Sajan Saini UIET Panjab University Chandigarh-160014 INDIA Department of Mathematics Panjab University Chandigarh-160014 INDIA Maharaja Ranjit Singh Punjab Technical University Bhatinda-15001 INDIA Post Graduate Institute of Medical Education and Research Chandigarh-160012 INDIA

Non-invasive detection of anemia is generally done by physical examination of regions like palpebral conjunctiva, fingernails, tongue and palmar creases. However, such examination is subject to large inter- and intra-observer bias. This problem can be alleviated by automating the anemia detection process through computerized analysis of images of these regions. The automated process includes sub-processes like preprocessing, segmentation of region-of-interest (ROI), ROI analysis or feature extraction, and classification. Of all these sub-processes, segmentation is the most crucial one as it helps in the precise extraction of ROI where the most crucial information for decision making lies. Recently, deep learning-based architectures have given exemplary performance in biomedical image segmentation. This paper is unique in a sense that it simulatneously analyzes performance of five deep learning-based architectures namely UNet, UNet++, FCN, PSPNet, and LinkNet. The experiments are performed on customly built dataset comprising of 2592 palpebral images of the pediatric population. The experimental results indicate that as compared to its counterparts, the LinkNet architecture performs the best. Its scores 94.17%, 90.14% and 93.78% for the accuracy, intersection-over-union (IoU), Dice score performance metrics, respectively. The study concludes that LinkNet architecture can be used for real-time segmentation of palpebral conjunctiva from images.

关键词： Medical Imaging image Segmentation deep learning image processing Anemia Detection Haemoglobin Palpebral Conjunctiva

来源：评论

学校读者我要写书评

暂无评论

Study on the Recognition of Visible image at Sea Based on YOLOv2-Network Model

Study on the Recognition of Visible Image at Sea Based on YO...

引用

International Conference on Guidance, Navigation and Control, ICGNC 2020

作者： Wang, Yifan Zhou, Haibin Zhang, Wenyi Luo, Fuyu System Engineering Research Institute Beijing100094 China

ISBN: (纸本)9789811581540

With the continuous maturity and improvement of machine learning, an intelligent maritime monitoring system is constructed for the identification and tracking of dynamic fuzzy targets in maritime video images. After studying the characteristics of the Marine target environment, it is found that the imaging effect is poor due to the violent movement of the offshore imaging platform, and the field of view of the ship is relatively small, so it is difficult to extract the characteristic information of the corresponding target. Depending on these reasons above, under the CAFFE framework of deep learning, the original and improved yolov2-a network model, SVM support vector machine, HOG feature, multi-scale transformation and multi-thread technology are adopted to identify multiple targets. The fast and effective intelligent automatic detection and recognition algorithm for fuzzy image is completed, which supports a variety of standards and definition. The final research results can detect the weak floating, small boats, speedboats, cruise ships and other targets, the identification accuracy reached more than 95%, for hd video image, real-time up to 30 frames/second. © 2022, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

YOLOv8-Powered Multi-Object Detection for Safety Helmets and Number Plates in real-time Surveillance Systems

YOLOv8-Powered Multi-Object Detection for Safety Helmets and...

引用

Communication, Computing and Industry 4.0 (C2I4), International Conference on

作者： Pramila S Anil D O Sirisha New Horizon College of Engineering Bengaluru India Department of CSE UVCE Bangalore University Bengaluru India Department of Computer Science and Business Systems G.Pulla Reddy Engineering College Kurnool India

ISBN: (数字)9798350379884

ISBN: (纸本)9798350379891

This paper presents a real-time object detection system using the latest YOLOv8 model to enhance traffic monitoring and safety enforcement. The model was trained to detect helmets, riders and number plates in diverse traffic scenarios, achieving an overall accuracy of 93.4%. Significant improvements were observed in precision, recall and mean Average Precision (mAP) scores due to the combination of an optimized architecture, high-quality data and fine-tuned hyperparameters. The system maintains real-time inference capabilities with a processing speed of 11 ms per image, making it suitable for continuous surveillance and enforcement applications. This study demonstrates the potential of using advanced deep learning techniques for effective and automated traffic rule enforcement. Future work may involve further optimizations and integration with large-scale cloud-based systems for more comprehensive deployments.

关键词： Accuracy Surveillance Transfer learning Object detection Traffic control Helmets real-time systems Safety License plate recognition Optimization

来源：评论

学校读者我要写书评

暂无评论

Predicting Room Impulse Responses Through Encoder-Decoder Convolutional Neural Networks 33

Predicting Room Impulse Responses Through Encoder-Decoder Co...

引用

33rd IEEE International Workshop on Machine learning for Signal processing, MLSP 2023

作者： Martin, I. Pastor, F. Fuentes-Hurtado, F. Belloch, J.A. Azpicueta-Ruiz, L. Naranjo, V. Pinero, G. Universidad Carlos Iii de Madrid Dep. Tecnología Electrónica Spain I3B Universitat Politècnica de València Spain Universidad Politécnica de Madrid Dep. Sistemas Informáticos Spain Universidad Politécnica de Madrid Knodis Research Group Spain Universidad Carlos Iii de Madrid Dep. Teoría de la Señal y Comunicaciones Spain Iteam Universitat Politécnica de València Spain

ISBN: (纸本)9798350324112

This paper investigates the ability of deep neural networks (DNNs) to predict a specific room impulse response (RIR) between two given room locations. We use three end-to-end deep learning (DL) models based on the encoder-decoder structure: an auto-encoder (AE), a variational AE (VAE) and a UNet. They try to generate a new RIR given: 1) the short-time Fourier transform (STFT) of a true RIR from the same room, 2) the spatial coordinates of the true and new RIRs, and 3) several room-related parameters. On the one hand, the magnitude and phase of the STFT are computed and presented to the DNN input. On the other hand, the spatial coordinates and the room parameters form an information vector that is embedded into the DNNs through their latent space (AE, VAE) or equivalent layer (UNet). A real database of RIRs measured in five different rooms was used to train and test the models. Two experiments were carried out to study the influence of the magnitude and phase terms of the loss function on the performance of the models. An additional experiment investigated the ability of the DL models to generalize across rooms. Results show that the three DNNs are able to predict the magnitude of the STFT, but they cannot accurately predict its phase. When comparing their ability to generalize across spaces, the UNet achieves the best results. © 2023 IEEE.

关键词： deep neural networks

来源：评论

学校读者我要写书评

暂无评论

ML-SIM: universal reconstruction of structured illumination microscopy images using transfer learning

引用

BIOMEDICAL OPTICS EXPRESS 2021年第5期12卷 2720-2733页

作者： Christensen, Charles N. Ward, Edward N. Lu, Meng Lio, Pietro Kaminski, Clemens F. Univ Cambridge Dept Chem Engn & Biotechnol Laser Analyt Grp Philippa Fawcett Dr Cambridge England Univ Cambridge Dept Comp Sci & Technol Artificial Intelligence Grp JJ Thomson Ave Cambridge England

Structured illumination microscopy (SIM) has become an important technique for optical super-resolution imaging because it allows a doubling of image resolution at speeds compatible with live-cell imaging. However, the reconstruction of SIM images is often slow, prone to artefacts, and requires multiple parameter adjustments to reflect different hardware or experimental conditions. Here, we introduce a versatile reconstruction method, ML-SIM, which makes use of transfer learning to obtain a parameter-free model that generalises beyond the task of reconstructing data recorded by a specific imaging system for a specific sample type. We demonstrate the generality of the model and the high quality of the obtained reconstructions by application of ML-SIM on raw data obtained for multiple sample types acquired on distinct SIM microscopes. ML-SIM is an end-to-end deep residual neural network that is trained on an auxiliary domain consisting of simulated images, but is transferable to the target task of reconstructing experimental SIM images. By generating the training data to reflect challenging imaging conditions encountered in real systems, ML-SIM becomes robust to noise and irregularities in the illumination patterns of the raw SIM input frames. Since ML-SIM does not require the acquisition of experimental training data, the method can be efficiently adapted to any specific experimental SIM implementation. We compare the reconstruction quality enabled by ML-SIM with current state-of-the-art SIM reconstruction methods and demonstrate advantages in terms of generality and robustness to noise for both simulated and experimental inputs, thus making ML-SIM a useful alternative to traditional methods for challenging imaging conditions. Additionally, reconstruction of a SIM stack is accomplished in less than 200 ms on a modern graphics processing unit, enabling future applications for real-time imaging. Source code and ready-to-use software for the method are available at http

关键词： image quality image resolution Imaging systems Live cell imaging Low light levels real time imaging

来源：评论

学校读者我要写书评

暂无评论

Rainfall Forecasting Using deep learning and Cloud image Analysis

Rainfall Forecasting Using Deep Learning and Cloud Image Ana...

引用

Electronics and Sustainable Communication Systems (ICESC), 2020 International Conference on

作者： K. RamaKrishna Reddy V. Sravya Reddy V. Sriram Sathvik Shaik Mahaboob Subhani Department of Computer Science and Engineering(AI & ML) Vignan Institute of Technology and Science

ISBN: (数字)9798350379945

ISBN: (纸本)9798350379952

This study introduces a new approach to rainfall prediction using deep learning and Convolutional Neural Networks (CNNs) for analyzing cloud image data sourced from diverse platforms. Traditional methods often struggle to capture precipitation patterns' complexity, necessitating advanced technologies like CNNs. The methodology emphasizes meticulous data preprocessing, including multiple discriminations, ratio adjustment, and augmentation, optimizing input data for CNNs to enhance quality, robustness, and predictive model effectiveness. A binary classification component enhances prediction accuracy and provides valuable insights into rainfall occurrence's binary nature. Using the multi-class weather dataset (MWD), this study explores weather image classification, examining dataset structure and showcasing sample images for each weather type. Two models with distinct architectures are trained from scratch, alongside a transfer learning model utilizing the VGG 16 pre-trained model. Result demonstrate promising performance across all models on the test dataset, highlighting their ability to accurately classify weather types. This study contributes to advancing rainfall prediction methodologies, offering insights into CNN application for weather analysis and forecasting, addressing critical needs in various sectors reliant on accurate weather predictions.

关键词： deep learning Rain Accuracy Transfer learning Predictive models real-time systems Convolutional neural networks Forecasting Meteorology image classification

来源：评论

学校读者我要写书评

暂无评论

Fast Monocular Visual Place Recognition for Non-Uniform Vehicle Speed and Varying Lighting Environment

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2021年第3期22卷 1679-1696页

作者： Li, Chu-Tak Siu, Wan-Chi Hong Kong Polytech Univ Dept Elect & Informat Engn Hong Kong Peoples R China

This paper presents a novel Fast Monocular Visual Place Recognition (FMPR) with a shallow path-oriented offline learning stage and an online place recognition and tracking stage. FMPR uses a tube of frames with a humanlike key frame recognition to solve place recognition for situations with varying speeds and changing lighting conditions, which are two most commonly encountered situations in real life. We propose an offline learning to analyze the correlation of all video frames in a reference path and to extract effective feature patches of key frames with an offline feature-shifts approach to achieve real-time place recognition. Our recognition results are on the basis of both the instant feature matching of frames and the historical recognition results which impose temporal logic constraints on the movement of a vehicle. Experimental results demonstrate that our proposed method can achieve comparable or even better performance compared with the state-of-the-art methods on different challenging datasets, especially for the case which requires a trade-off between the performance and the processing time. We believe that our FMPR offers a useful alternative to computationally expensive deep learning-based methods especially for applications with battery-powered or resource-limited devices.

关键词： Visualization Feature extraction Lighting image recognition Cameras Databases Electron tubes Autonomous driving key frame identification place recognition and tracking vehicle detection

来源：评论

学校读者我要写书评

暂无评论

Quality Inspection and Defect Recognition Technology for Intelligent Manufacturing Based on Machine Vision

Quality Inspection and Defect Recognition Technology for Int...

引用

Digital Analysis and processing, Intelligent Computation (DAPIC), International Conference on

作者： Xiaojian Lin Wen Huang Harbin Huade University Harbin Heilongjiang China

ISBN: (数字)9798331536169

ISBN: (纸本)9798331536176

With the rapid development of intelligent manufacturing, quality inspection and defect recognition have become one of the key technologies to improve production efficiency and product quality. Traditional manual inspection methods cannot meet the demands of efficient, accurate and real-time production, so the quality inspection and defect recognition technology based on machine vision has gradually become an important application in the field of intelligent manufacturing. This paper firstly introduces the basic principles of machine vision technology and the background of its application in intelligent manufacturing, and then discusses in detail the key technologies in quality inspection and defect recognition, including image acquisition and pre-processing, feature extraction and matching algorithms, as well as defect classification and determination methods. This paper also analyzes the design and implementation of the intelligent manufacturing quality inspection system, including the system architecture, data processing flow and practical application cases. An outlook is given for the challenges faced by the current technology and the future development trend. This paper aims to provide reference for researchers and engineers in related fields to promote the further development and application of machine vision technology in intelligent manufacturing.

关键词： deep learning Manufacturing industries Machine vision Production Inspection Feature extraction real-time systems Product design Manufacturing Quality assessment

来源：评论

学校读者我要写书评

暂无评论

3rd International Conference on Recent Trends in image processing and Pattern Recognition, RTIP2R 2020

3rd International Conference on Recent Trends in Image Proce...

引用

3rd International Conference on Recent Trends in image processing and Pattern Recognition, RTIP2R 2020

ISBN: (纸本)9789811605062

The proceedings contain 78 papers. The special focus in this conference is on Recent Trends in image processing and Pattern Recognition. The topics include: Ensemble of Nested Dichotomies for Author Identification System Using Similarity-Based Textual Features;Feature Combination of Pauli and H/A/Alpha Decomposition for Improved Oil Spill Detection Using SAR;a Fast and Efficient Convolutional Neural Network for Fruit Recognition and Classification;copy-Move image Forgery Detection Using Discrete Wavelet Transform;a Comprehensive Survey of Different Phases for Involuntary System for Face Emotion Recognition;classification of Vehicle Type on Indian Road Scene Based on deep learning;indian Road Lanes Detection Based on Regression and clustering using Video processing Techniques;detection of Emotion Intensity Using Face Recognition;double Authentication System Based on Face Identification and Lipreading;fuzzy Approach to Evaluate Performance of Teaching Staff in Technical Institutions;safety Gear Check at Industries and Laboratories Using Convolutional Neural Network Based on deep learning;analysis of Changing Trends in Textual Data Representation;detection of Falsary Happening on Social Media Using image processing: Feature Extraction and Matching;Development of Multi Faces Recognition System Using HOG Features and Neural Network Classifier in real time Environment;extraction of Key Frame from Random Videos Based On Discrete Cosine Transformation;Prediction of SO2 Air Pollution Quality Parameter of Kolhapur City Using time Series Analysis;A Big Data Prediction for Weather Forecast Using Hybrid ARIMA-ANN time Series Model;automatic Detection of Riots Using deep learning;protecting Big Data Sets from Unauthorized Users on Cloud;text Categorization: A Lazy learning-Based Approach.

关键词：

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：