检索结果-内蒙古大学图书馆

PLSAV: Parallel loop searching and verifying for loop closure detection

IET INTELLIGENT TRANSPORT SYSTEMS 2021年第5期15卷 683-698页

作者： Yang, Zhe Pan, Yun Deng, Lei Xie, Yuan Huan, Ruohong Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou Peoples R China Univ Calif Santa Barbara Dept Elect & Comp Engn Santa Barbara CA 93106 USA Zhejiang Univ Technol Coll Comp Sci & Technol Hangzhou Peoples R China

Visual simultaneous localization and mapping (vSLAM), one of the most important applications in autonomous vehicles and robots to estimate the position and pose using inexpensive visual sensors, suffers from error accumulation for long-term navigation without loop closure detection. Recently, deep neural networks (DNNs) are leveraged to achieve high accuracy for loop closure detection, however the execution time is much slower than those employing handcrafted visual features. In this paper, a parallel loop searching and verifying method for loop closure detection with both high accuracy and high speed, which combines two parallel tasks using handcrafted and DNN features, respectively, is proposed. A fast loop searching is proposed to link the bag-of-words features and histogram for higher accuracy, and it splits the images into multiple grids for high parallelism;meanwhile, a DNN feature extractor is utilized for further verification. A loop state control method based on a finite state machine to control these tasks is designed, wherein the loop closure detection is described as a context-related procedure. The framework is implemented on a real machine, and the top-2 best accuracy and fastest execution time of 80-543 frames per second (min: 1.84ms, and max: 12.45ms) are achieved on several public benchmarks compared with some existing algorithms.

关键词： image recognition Mobile robots Automata theory Computer vision and image processing techniques Machine learning (artificial intelligence) Neural nets

来源：评论

学校读者我要写书评

暂无评论

Bimodal Feature Propagation and Fusion for real-time Semantic Segmentation on RGB-D images 7

Bimodal Feature Propagation and Fusion for Real-time Semanti...

引用

7th International Conference on Intelligent Computing and Signal processing, ICSP 2022

作者： He, Ying Xiao, Li Sun, Zhigang Wang, Zhuo Huazhong University of Science and Technology Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Wuhan China

ISBN: (纸本)9781665478571

Semantic Segmentation is the foundation of scene understanding and automatic driving tasks. One of the challenges of semantic segmentation is the reduction of feature resolution as the network goes deep. In this paper, the low-resolution features are integrated into high-resolution features progressively, enabling high-resolution prediction. Currently, incorporating the auxiliary depth information into a semantic segmentation framework has proven to be helpful to improve accuracy. However, additional process of depth data immensely increases computational complexity, resulting in limitation for practical application. Here, a real-time bimodal semantic segmentation network that effectively extracts and fuses complementary information is investigated. Our network extends a light-weight architecture which can seamlessly take advantage of various prevalent backbone networks. Our model extracts RGB and depth features from two branches in parallel and integrates multi-level features of two modalities through bimodal feature fusion blocks. To reduce computation complexity, we employ depth-wise separable convolution while keep performance intact. Meanwhile, an attention mechanism is exploited to recalibrate the fusion modality features adaptively. Refinement components are retained to refine multi-resolution feature maps. We conduct ablative study to demonstrate the availability of our bimodal fusion blocks. Comprehensive experiments validate that the proposed network outperforms the state-of-the-arts and achieves 18 FPS on the challenging RGB-D dataset, NYUDv2. © 2022 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Design of Automatic Lung Nodule Detection System Based on Multi-Scene deep learning Framework

引用

IEEE ACCESS 2020年 8卷 90380-90389页

作者： Zhang, Qinghai Kong, Xiaojing Third Peoples Hosp Zibo Imaging Dept Zibo 255000 Peoples R China

Nowadays, the efficient identification of the lung nodule greatly leads to the chance of lung cancer risk assessment. Hence, the exact locations of lung nodules are a critical and complicated task. Researchers in this area have been working widely for almost two years. However, previous computer-aided detection (CAD) modules, such as transforming CT, segmenting the lung nodule and extracting the features are mostly complex and time-consuming, because more modules will require the creation of a complete image processing system. In addition, certain state-of-the-art deep learning systems are specified in the database standard. For this purpose, this paper suggests an efficient identification system for lung nodules based on Multi-Scene deep learning Framework (MSDLF) by the vesselness filter. A four-channel CNN model is designed to enhance the radiologist & x2019;s knowledge in the detection of four-stage nodules by integrating two image Scenes. This model can be applied in two different classes. The results show that the Multi-Scene deep learning Framework (MSDLF) is efficient for increasing the accuracy and significantly reducing false positives in an enormous amount of image data in the detection of lung nodules.

关键词： Lung Computed tomography Cancer Machine learning image segmentation Databases Multi-scene deep learning framework vesselness filter convolutional neural network (CNN) lung nodules

来源：评论

学校读者我要写书评

暂无评论

Sensor Fusion in Autonomous Vehicle with Traffic Surveillance Camera System: Detection, Localization, and AI Networking

引用

SENSORS 2023年第6期23卷 3335页

作者： Hasanujjaman, Muhammad Chowdhury, Mostafa Zaman Jang, Yeong Min Khulna Univ Engn & Technol KUET Dept Elect & Elect Engn Khulna 9203 Bangladesh Kookmin Univ Dept Elect Engn Seoul 02707 South Korea

Complete autonomous systems such as self-driving cars to ensure the high reliability and safety of humans need the most efficient combination of four-dimensional (4D) detection, exact localization, and artificial intelligent (AI) networking to establish a fully automated smart transportation system. At present, multiple integrated sensors such as light detection and ranging (LiDAR), radio detection and ranging (RADAR), and car cameras are frequently used for object detection and localization in the conventional autonomous transportation system. Moreover, the global positioning system (GPS) is used for the positioning of autonomous vehicles (AV). These individual systems' detection, localization, and positioning efficiency are insufficient for AV systems. In addition, they do not have any reliable networking system for self-driving cars carrying us and goods on the road. Although the sensor fusion technology of car sensors came up with good efficiency for detection and location, the proposed convolutional neural networking approach will assist to achieve a higher accuracy of 4D detection, precise localization, and real-time positioning. Moreover, this work will establish a strong AI network for AV far monitoring and data transmission systems. The proposed networking system efficiency remains the same on under-sky highways as well in various tunnel roads where GPS does not work properly. For the first time, modified traffic surveillance cameras have been exploited in this conceptual paper as an external image source for AV and anchor sensing nodes to complete AI networking transportation systems. This work approaches a model that solves AVs' fundamental detection, localization, positioning, and networking challenges with advanced image processing, sensor fusion, feathers matching, and AI networking technology. This paper also provides an experienced AI driver concept for a smart transportation system with deep learning technology.

关键词： autonomous vehicle AI networking deep learning localization positioning sensor fusion traffic surveillance camera

来源：评论

学校读者我要写书评

暂无评论

Joint Parameterized Resource Allocation for IAB Networks with deep Reinforcement learning

Joint Parameterized Resource Allocation for IAB Networks wit...

引用

International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： Hou, Xinwu Huang, Yihang Xu, Yin He, Dazhi Zhang, Wenjun Shanghai Jiao Tong Univ Cooperat Media Network Innovat Ctr CMIC Sch Elect Informat & Elect Engn Informat & Elect Engn Shanghai Peoples R China

ISBN: (纸本)9781665469012

The integrated access and backhaul (IAB) demonstrates significant advantages in terms of economy and stability over traditional fiber-like links. Although promising, with the high difficulty of interference management, appropriately allocating the spectrum bandwidth and power resources in IAB system to each communication link is still challenging. In this paper, we investigate the joint optimization problem of spectrum allocation and power management in an IAB network operating in the mm-Wave, which aims to improve the total date rate of the IAB system while satisfying the users' quality of service (QoS) requirements and to ensure fairness among users with the same QoS requirements. With these goals, we formulate the joint optimization as a mixed integer nonlinear programming (MINLP) problem. To solve this problem, we propose to use a state-of-the-art actor-critic based reinforcement learning algorithm in the parameterized action space, namely parameterized action-space deep deterministic policy gradient (PADDPG), to obtain an efficient optimization strategy on bandwidth and power allocation, called parameterized resource allocation (PRA). The proposed scheme can achieve model-free, real-time resources allocation in the absence of perfect CSI. Numerical experiments demonstrate the superiority of our scheme over benchmark algorithms in terms of data rates, QoS satisfaction and user fairness.

关键词： Integrated access and backhaul PADDPG PRA capacity QoS fairness

来源：评论

学校读者我要写书评

暂无评论

real-time Vehicle Detection and Classification on the Padma Multipurpose Bridge in Bangladesh Using a deep learning Model

Real-time Vehicle Detection and Classification on the Padma ...

引用

Next-Generation Computing, IoT and Machine learning (NCIM), International Conference on

作者： Md. Alamgir Hossain Md. Atikur Rahman Rukaia Khanom Sahelee Sultana Amran Hossain Department of Computer Science and Engineering Dhaka University of Engineering & Technology Gazipur Gazipur Bangladesh

Rapid and accurate recognition and classification of vehicles are essential for the Intelligent Transportation System (ITS) of Bangladesh. The paper presents a deep learning-based object detection model called Vehicle-YOLOv7 that can detect and classify moving vehicles in real-time on the Padma Multipurpose Bridge (PMB) in Bangladesh. Initially, six hundred-fifty-five (655) still pictures and a video frame containing vehicle images: Bus, Microbus, Car, Truck, Minitruck, Pickup, and Trailer are collected from the PMB to create an original PMB-vehicle image dataset. Thereafter, pre-processing techniques and augmentation processes were employed to produce a training dataset of 7205 image samples. The model was trained and tested using 80% and 20% of the images respectively. Two variations of the YOLOv7 model were evaluated. The Vehicle-YOLOv7 (i.e., YOLOv7x) model performed better than the YOLOv7 model and other state-of-the-art vehicle identification models. The proposed model attained accuracy (A), precision (P), Recall (R), F1-score, and mean-average-precision (mAP) of 98.12%, 97.20%, 98.0%, 96.80%, and 96.86%, respectively. The model accurately detected vehicles with a bounding box and showed detection score and classified the moving vehicles into eight classes: Bus, Microbus, Car, Truck, Minitruck, Motorcycle, Pickup, and Trailer. In addition, the proposed model automatically calculated the vehicles' toll, which reduces the complexity of toll collection by ITS. Therefore, the Vehicle-YOLOv7 detection model is a trustworthy method that can be used to detect and classify the vehicles in the ITS of Bangladesh.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Abnormal Sensor Signal Detection With Channelwise Reconstruction Errors

引用

IEEE ACCESS 2021年 9卷 39995-40007页

作者： Kwak, Mingu Kim, Seoung Bum Korea Univ Sch Ind Management Engn Seoul 02841 South Korea

Detecting an anomaly in multichannel signal data is a challenging task in various domains. It should take into account the cross-channel relationship and temporal relationship within each channel. Moreover, the signal data is high-dimensional and making it difficult to gather sufficient abnormal labels. Consequently, unsupervised reconstruction-based anomaly detection methods have been applied successfully in many studies. However, they lose valuable channel information inherent in the reconstruction errors by merely averaging the errors for both the channel and time, then consider the average value as an anomaly score. In this study, we propose a method to explicitly employ channelwise reconstruction errors as a feature to detect abnormal signals. After a convolutional autoencoder produces the channelwise reconstruction errors, a machine learning anomaly detection model aggregates the errors as an anomaly score. To demonstrate the effectiveness and applicability of the proposed model, we conduct experiments using simulated data and real-world automobile data. The results show that the proposed method remarkably enhances the detectability compared to the simple average of the reconstruction errors. The reconstruction errors of abnormal and normal channels are shown to be different;therefore, it can be considered as an appropriate feature for anomaly detection. The best performance is obtained by using local outlier factors in the following anomaly detection model.

关键词： Anomaly detection image reconstruction Feature extraction Data models time series analysis Principal component analysis Convolution Anomaly detection convolutional autoencoder deep learning multichannel sensor signal data unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Leveraging retinanet based object detection model for assisting visually impaired individuals with metaheuristic optimization algorithm

引用

Scientific Reports 2025年第1期15卷 1-19页

作者： Khadidos, Alaa O. Yafoz, Ayman Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah Saudi Arabia King Salman Center for Disability Research Riyadh 11614 Saudi Arabia

Visually impaired individuals suffer many problems in handling their everyday activities like road crossing, writing, finding an object, reading, and so on. However, many navigation methods are available, and efficient object detection (OD) methods for visually impaired people (VIP) need to be improved. OD is the most crucial contribution of computer vision (CV) and plays an important part in recognizing and finding objects in the image. The aged and visually challenged people can identify different objects accurately under some constraint features like scaled, occlusion, illuminated, and blurred nature in different real world. A considerable number of studies are performed in the domain of real-time object recognition using deep learning (DL). The DL-based methods remove the features autonomously and classify and detect the objects. This paper proposes a novel Object Detection Model for Visually Impaired Individuals with a Metaheuristic Optimization Algorithm (ODMVII-MOA) technique. The proposed ODMVII-MOA technique aims to improve OD methods in real-time with advanced techniques to detect and recognize objects for disabled people. At first, the image pre-processing stage applies the Weiner filter (WF) method to enhance image quality by eliminating the unwanted noise from the data. Furthermore, the RetinaNet technique is utilized for the OD process to recognize and locate objects within an image. Besides, the proposed ODMVII-MOA method employs the EfficientNetB0 method for the feature extraction process. For the classification process, the LSTM-AE method is employed. Finally, the Dandelion Optimizer (DO) method adjusts the hyperparameter range of the LSTM-AE method optimally and results in better performance of classification. The experimental validation of the ODMVII-MOA model is verified under the indoor OD dataset and the outcomes are determined regarding different measures. The comparison study of the ODMVII-MOA model portrayed a superior accuracy value of 99.6

关键词： Computer vision Dandelion optimizer deep learning Feature extraction Object detection Visually impaired individuals

来源：评论

学校读者我要写书评

暂无评论

Indoor Space Flow Analysis Based on deep learning

Indoor Space Flow Analysis Based on Deep Learning

引用

International Conference on Artificial Intelligence in Information and Communication (ICAIIC)

作者： Chang Woo Choi Hyo Eun Kang Yoon Young Hong Yong Su Kim Guem Bo Kim Aji Teguh Prihatno Jang Hyun Ji Seung Do Hong Ho Won Kim Dept. of Information and Computer Engineering Pusan National University Busan South Korea Smartm2m Pusan National University Busan South Korea Air Solution R&D Laboratory LG Electronics Changwon South Korea

It is essential to perform flow analysis in all spaces where people live. For example, designing the shape of the wing by analyzing the flow flowing through the wing of an airplane, or finding an appropriate air conditioner installation location by analyzing the flow according to the location of the air conditioner in the indoor space. In this study, we propose a deep learning model that performs real-time flow analysis assuming an indoor space that is relatively smaller than outdoor space. Computational Fluid Dynamics (CFD), a traditional method used for flow analysis, is not suitable for this task because it takes a long time to derive simulation results. Thus, the application of deep learning to flow analysis is considered in the present study because deep learning technology for physics, i.e., fluid mechanics and thermodynamics, can be applied to real spaces. We have constructed a deep learning model based on the TransUnet model that can learn data relationships and capture spatial information. Unlike the existing TransUnet model, our model contains a dense layer to reflect operating and spatial information. train and test data were collected using the ANSYS FLUENT commercial program. On 11 test data cases, the average R2 score between the actual and predicted value was 0.884, and the RMSE was 0.047, which are significant results. We used the image of the entire space as well as a cross-section to see how similar the predicted values were to the actual ones, Although a slight error occurred inside the space, It was confirmed that the flow tendency was accurately learned under the given operating conditions. Flow analysis through simulation based on existing numerical analysis methods requires a minimum of 8 hours for processing. However, our proposed deep learning model significantly reduces the time cost of flow analysis as it requires less than 3 seconds.

关键词： deep learning Analytical models Solid modeling Three-dimensional displays Fluids Costs Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

Disaster Management Ecosystem Using Crowdsourced Data, deep learning and NLP

Disaster Management Ecosystem Using Crowdsourced Data, Deep ...

引用

Sustainable Communication Networks and Application (ICSCNA), International Conference on

作者： Vaishali Savale Siddhi Patil Atharva Raut Mrunmayee Phadke Rahul Sundkar Rajkumar Dongre Department of Artificial Intelligence and Data Science Engineering Vishwakarma Institute of Technology Pune Maharashtra India

ISBN: (数字)9798331530013

ISBN: (纸本)9798331530020

In recent years, the frequency and intensity of natural disasters have increased, necessitating more efficient and responsive disaster management systems. This paper presents a novel disaster management solution leveraging crowdsourced data to enhance disaster response and coordination. The system bridges the gap between citizens and emergency management officials by composing a mobile application for citizens to send real-time disaster alerts and a web application for administrators to monitor and manage these alerts. A customized DBSCAN algorithm, adapted to use haversine distance, is deployed to detect clusters of disaster alerts in real time, indicating potential disaster events. Upon cluster detection, administrators can delineate affected areas on a map, triggering notifications to users within these regions and enabling them to send help requests with images and messages. These help requests are processed using the BLIP model for image captioning and the ChatGPT API for categorization, assigning relevant tags to each request. The system provides administrators with real-time data analysis features, including map filtering and citizen help request category visualization, enabling more informed decision-making. Utilizing a cloud-based architecture ensures scalability, reliability, and efficient processing of large volumes of data. This research demonstrates the system's capability to provide timely alerts and comprehensive situational awareness, ultimately contributing to more efficient disaster management.

关键词： Cloud computing Disasters Scalability Decision making Disaster management Computer architecture Chatbots real-time systems Reliability Monitoring

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：