检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Jiang, Chaokang Wang, Guangming Wu, Jinxing Miao, Yanzi Wang, Hesheng Engineering Research Center of Intelligent Control for Underground Space Ministry of Education School of Information and Control Engineering Advanced Robotics Research Center China University of Mining and Technology Xuzhou221116 China Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Shanghai200240 China The Department of Engineering Mechanics Shanghai Jiao Tong University Shanghai200240 China

Promising complementarity exists between the texture features of color images and the geometric information of LiDAR point clouds. However, there still present many challenges for efficient and robust feature fusion in the field of 3D object detection. In this paper, first, unstructured 3D point clouds are filled in the 2D plane and 3D point cloud features are extracted faster using projection-aware convolution layers. Further, the corresponding indexes between different sensor signals are established in advance in the data preprocessing, which enables faster cross-modal feature fusion. To address LiDAR points and image pixels misalignment problems, two new plug- and-play fusion modules, LiCamFuse and BiLiCamFuse, are proposed. In LiCamFuse, soft query weights with perceiving the Euclidean distance of bimodal features are proposed. In BiLiCamFuse, the fusion module with dual attention is proposed to deeply correlate the geometric and textural features of the scene. The quantitative results on the KITTI dataset demonstrate that the proposed method achieves better feature-level fusion. In addition, the proposed network shows a shorter running time compared to existing methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Representative Graph Neural Network 16th

Representative Graph Neural Network

引用

16th European Conference on Computer Vision, ECCV 2020

作者： Yu, Changqian Liu, Yifan Gao, Changxin Shen, Chunhua Sang, Nong Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China The University of Adelaide Adelaide Australia

ISBN: (纸本)9783030585709

Non-local operation is widely explored to model the long-range dependencies. However, the redundant computation in this operation leads to a prohibitive complexity. In this paper, we present a Representative Graph (RepGraph) layer to dynamically sample a few representative features, which dramatically reduces redundancy. Instead of propagating the messages from all positions, our RepGraph layer computes the response of one node merely with a few representative nodes. The locations of representative nodes come from a learned spatial offset matrix. The RepGraph layer is flexible to integrate into many visual architectures and combine with other operations. With the application of semantic segmentation, without any bells and whistles, our RepGraph network can compete or perform favourably against the state-of-the-art methods on three challenging benchmarks: ADE20K, Cityscapes, and PASCAL-Context datasets. In the task of object detection, our RepGraph layer can also improve the performance on the COCO dataset compared to the non-local operation. Code is available at https://***/RepGraph. © 2020, Springer Nature Switzerland AG.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Plant Phenotyping by Combining Domain Adaptation with 3D Plant Model Simulations: Application to Wheat Leaf Counting at Seedling Stage

引用

Plant Phenomics 2023年第2期5卷 226-238页

作者： Yinglun Li Xiaohai Zhan Shouyang Liu Hao Lu Ruibo Jiang Wei Guo Scott Chapman Yufeng Ge Benoit de Solan Yanfeng Ding Frédéric Baret Plant Phenomics Research Centre Academy for Advanced Interdisciplinary StudiesJiangsu Collaborative Innovation Center for Modern Crop ProductionNanjing Agricultural UniversityNanjingChina Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and AutomationHuazhong University of Science and TechnologyWuhanChina Graduate School of Agricultural and Life Sciences The University of Tokyo1-1-1 Midori-choNishitokyo CityTokyoJapan School of Agriculture and Food Sciences The University of QueenslandSt.LuciaQueensland 4072Australia Department of Biological Systems Engineering University of Nebraska-LincolnLincolnNebraska 68583United States INRAE Avignon UniversitéUMR EMMAHUMT CAPTE228route de l’aérodrome-CS 4050984914 Avignon Cedex 9France ARVALIS Institut du Végétal 3 Rue Joseph et Marie Hackin75116 ParisFrance

The number of leaves at a given time is important to characterize plant growth and *** this work,we developed a high-throughput method to count the number of leaves by detecting leaf tips in RGB *** digital plant phenotyping platform was used to simulate a large and diverse dataset of RGB images and corresponding leaf tip labels of wheat plants at seedling stages(150,000 images with over 2 million labels).The realism of the images was then improved using domain adaptation methods before training deep learning *** results demonstrate the efficiency of the proposed method evaluated on a diverse test dataset,collecting measurements from 5 countries obtained under different environments,growth stages,and lighting conditions with different cameras(450 images with over 2,162 labels).Among the 6 combinations of deep learning models and domain adaptation techniques,the Faster-RCNN model with cycle-consistent generative adversarial network adaptation technique provided the best performance(R^(2)=0.94,root mean square error=8.7).Complementary studies show that it is essential to simulate images with sufficient realism(background,leaf texture,and lighting conditions)before applying domain adaptation ***,the spatial resolution should be better than 0.6 mm per pixel to identify leaf *** method is claimed to be self-supervised since no manual labeling is required for model *** self-supervised phenotyping approach developed here offers great potential for addressing a wide range of plant phenotyping *** trained networks are available at https://***/YinglunLi/Wheat-leaf-tip-detection.

关键词： Wheat seedling Plant

来源：评论

学校读者我要写书评

暂无评论

Recurrent Attentional Reinforcement Learning for Fault Diagnosis of Hydraulic System

Recurrent Attentional Reinforcement Learning for Fault Diagn...

引用

International Conference on control, Decision and Information Technologies (CoDIT)

作者： Zhenhui Tang Ru Liu Kang Lou Chaojian Gao Jingcheng Wang SJTU Sanya Yazhou Bay Institute of Deepsea Science and Technology Sanya China Department of Automation Shanghai Jiao Tong University Shanghai China Shanghai Marine Equipment Research Institute Shanghai China Department of Automation the Key Laboratory of System Control and Information Processing Ministry of Education of China and the Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Shanghai China Autonomous Systems and Intelligent Control International Joint Research Center Xi'an Technological University Xi'an China

ISBN: (数字)9798350373974

ISBN: (纸本)9798350373981

Recognizing fault types of machinery system is a fundamental but challenging task in industrial application. Although data-driven fault diagnosis method attains remarkable progress by learning fault features automatically, the existing data-driven fault diagnosis pipelines learn non-discriminative feature. To deal with these problems, a new data-driven fault diagnosis method with recurrent attentional module is proposed to find a sequence of informative features by localizing distinguishing time fragments about each individual faults iteratively. The contextual sequence dependency between the localized time fragments is enhanced. Then, the fault types are further predicted on localized time fragments. The proposed framework also explicitly models long-term dependencies among these attentional time fragments to capture historical fault information. Experiments on hydraulic system show that the proposed framework achieves superior diagnosis performance and computation efficiency.

关键词： Fault diagnosis Location awareness Accuracy Computational modeling Pipelines Reinforcement learning Hydraulic systems Computational efficiency Information technology

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Based Scene Text Recognition for Industrial Applications

Knowledge-Based Scene Text Recognition for Industrial Applic...

引用

IEEE International Conference on Industrial Technology (ICIT)

作者： Guowei Deng Jingzheng Tu Cailian Chen Jianping He Xinyi Le Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai China Shanghai Engineering Research Center of Industrial Intelligent Control and Management Shanghai China

ISBN: (纸本)9781728119496

Scene text recognition (STR) methods combined with semantic information have made great progress to recognize texts in natural scenes, most of which are daily words. However, research on mining semantic information in industrial texts attracts less attention. Since industrial texts follow a different semantic pattern defined by industry standards, it challenges many existing methods to conduct accurate semantic reasoning. In this paper, we abstract the industry standards into two aspects of prior knowledge: the grouping property and the prior lexicon. Correspondingly, a knowledge-based language model is proposed with several group-wise correlation modules and a lexicon-based reasoning module to learn semantic rules from both data and the prior knowledge. Besides, we transfer the prior knowledge into data by generating synthetic pure text datasets according to the industry standards’ rules, which introduces more prior knowledge to the language model. Furthermore, a novel STR framework is presented by combining the knowledge-based language model and an attention-based vision model. For evaluation, two industrial text datasets called CIN and SB are collected from real-world industrial field surveillance. Experiments indicate that our method’s word-level accuracy outperforms state-of-the-art methods with 15% and 10.61% on CIN and SB datasets respectively.

关键词： Industries Correlation Text recognition Surveillance Semantics Knowledge based systems Data models

来源：评论

学校读者我要写书评

暂无评论

Atf-Bss: Adaptive Time-Frequency Blind Source Separation for High Reverberation and Echo Environments

SSRN

引用

SSRN 2023年

作者： Xie, Yuan Zou, Tao Chen, Ci Sun, Weijun Xie, Shengli School of Mechanical and Electrical Engineering Guangzhou University Guangdong-Hong Kong-Macao Key Laboratory of Multi-scale Information Fusion and Collaborative Optimization Control of Complex Manufacturing Process Guangzhou510006 China School of Automation Guangdong University of Technology Joint International Research Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education of the P.R.C. Guangzhou510006 China School of Automation Guangdong University of Technology Guangdong Key Laboratory of IoT Information Processing Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China School of Automation Guangdong University of Technology Key Laboratory of Intelligent Detection and the IoT in Manufacturing Ministry of Education 111 Center for Intelligent Batch Manufacturing Based on IoT Technology Guangzhou510006 China

Blind source separation with high reverberation and echoes is an open and extremely challenging issue. Room reverberation and echoes have an important influence on acoustics quality, causing great difficulty in independent source separation. In the paper, an adaptive time-frequency blind source separation algorithm (ATF-BSS) is designed to deal with high reverberation and echoes, which can adaptively attenuate audible echoes and enhance the harmonic structure of real-world audio signals, achieving excellent separation performance. Firstly, an equalizer network is designed for the room impulse response to weakening audible echoes under high reverberation and echo environments without changing the perceived timbre. Furthermore, an adaptive time-frequency Wiener-like masking is presented to enhance the dominant signal and attenuate the other components that have less energy than the dominant signal. To the best of our knowledge, this is the first attempt to incorporate the advantages of an equalizer network for removing or weakening audible echoes and adaptive time-frequency masking to solve blind separation with high reverberation and echoes. Experimental results on public audio datasets demonstrate that the proposed ATF-BSS algorithm improves signal intelligibility and achieves much better separation performance than state-of-the-art methods. © 2023, The Authors. All rights reserved.

关键词： Reverberation

来源：评论

学校读者我要写书评

暂无评论

Understanding the Robustness of 3D Object Detection with Bird'View Representations in Autonomous Driving

Understanding the Robustness of 3D Object Detection with Bir...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zijian Zhu Yichi Zhang Hai Chen Yinpeng Dong Shu Zhao Wenbo Ding Jiachen Zhong Shibao Zheng Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Dept. of Comp. Sci. and Tech. THBI Lab Institute for AI Tsinghua University BNRist Center Zhongguancun Laboratory Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Information Materials and Intelligent Sensing Laboratory of Anhui Province School of Computer Science and Technology Anhui University SAIC Motor AI Lab

3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations; 2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features; 3) Camera-LiDARfusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An advanced actor-critic algorithm for training video game ai 1st

An advanced actor-critic algorithm for training video game a...

引用

1st International Conference on Neural Computing for Advanced Applications, NCAA 2020

作者： Zha, Zhong Yi Tang, Xue Song Wang, Bo Key Laboratory of Ministry of Education for Image Processing and Intelligent Control Artificial Intelligence and Automation School Huazhong University of Science and Technology Wuhan430074 China College of Information and Science Donghua University Shanghai China

ISBN: (纸本)9789811576690

This paper presents an improved deep reinforcement learning (DRL) algorithm, namely Advanced Actor Critic (AAC), which is based on Actor Critic (AC) algorithm, to the video game Artificial intelligence (AI) training. The advantage distribution estimator, Normal Constraint (NC) function and exploration based on confidence are introduced to improve the AC algorithm, in order to achieve accurate value estimate, select continuous spatial action correctly, and explorate effectively. The aim is to improve the performance of conventional AC algorithm in a complex environment. A case study of video game StarCraft II mini-game AI training is employed. The results verify that the improved algorithm effectively improves the performance in terms of the convergence rate, maximum reward, average reward in every 100 episode, and time to reach a specific reward, etc. The analysis on how these modifications improve the performance is also given through interpretation of the feature layers in the mini-games. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Spatial-Temporal-Rule-based Traffic Movement Status Interpretation at Intersections *

Spatial-Temporal-Rule-based Traffic Movement Status Interpre...

引用

IEEE International Conference on Real-time Computing and Robotics (RCAR)

作者： Hanyang Zhuang Liyang Gao Yeqiang Qian Chunxiang Wang Ming Yang University of Michigan – Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University Shanghai China Department of Automation Shanghai Jiao Tong University Shanghai Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai Shanghai Engineering Research Center of Intelligent Control and Management Shanghai China

Traffic light is the most significant component of road transportation. Traffic lights control the traffic movement at intersections by allowing the movement of some lanes and forbidding that of the others to ensure safety and efficiency. It is important for intelligent vehicles to understand the meaning of traffic lights. Conventional research works on traffic light detection and tracking consider each light independently, so the occlusion of some lights may mislead the decision of the vehicle. Since all traffic lights at an intersection work together, it is more essential to understand the traffic movement status controlled by the traffic lights rather than the perception of a single light. Therefore, this paper proposes a spatial-temporal-rule-based framework to interpret the current traffic movement status and predict the next status. The spatial information, including position and class, is obtained by traffic light detection. The temporal information is perceived by the proposed topologic-feature-based tracking method. The rule information is recognized by combination of spatial and temporal information. A dual-view camera system consisting of a telephoto camera and a wide-angle camera is adopted to ensure long perception distance and wide coverage simultaneously. The successful tracking of traffic lights from 200 m to 10 m away provides enough data to recognize the phase-change rule, which leads to correct interpretation and prediction. A customized dataset was collected, and experiments were conducted to illustrate the capability of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Memristor-based multilayer neural network with edge learning and weight update

引用

Neurocomputing 2025年

作者： Ningye Jiang Jupeng Xie Mingxuan Jiang Haoen Huang Zhigang Zeng School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan 430074 China Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education of China Wuhan 430074 China

Memristor crossbar array (MCA) is a promising computing-in-memory architecture for accelerating vector-matrix multiplication (VMM). However, existing memristor-based circuits primarily perform forward computations without local learning capabilities. To address the issue, a memristor-based multilayer neural network (MNN) with edge learning is proposed, enabling error backpropagation and one-step weight update. In particular, the overall circuit design of a three-layer MNN is outlined, comprising three operational stages: forward computation, backpropagation, and weight update. Subsequently, the schematics of four adopted circuit blocks are detailed. By integrating the memory and weight update blocks with MCA, the proposed three-layer memristor-based MNN exhibits scalability and universality, allowing it to be extended to the deep neural network (DNN) architecture. Circuit simulations validate the effectiveness of the proposed scheme, while offline-online co-simulations demonstrate its strength in compensating for variations and non-ideal characteristics of physical devices. Comparative results further highlight the advantages of the proposed scheme in realizing full-analog edge learning circuits and performing error backpropagation processes efficiently.

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：