检索结果-内蒙古大学图书馆

Decoupling of nitrogen and oxygen sensors based on particle swarm optimization and fuzzy feedforward compensation

学校读者我要写书评

暂无评论

Decoupling of nitrogen and oxygen sensors based on particle ...

第43届中国控制会议

作者： Can Sun Zhen Wang Jie Wang Yu Qiao Xi Li School of Artificial Intelligence and Automation Key Laboratory of Image Processing and Intelligent Control of Education MinistryHuazhong University of Science and Technology Shenzhen Huazhong University of Science and Technology Research Institute

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

A nitrogen oxide sensor that can accurately and quickly detect the content of nitrogen oxides in exhaust gas plays a key role in the treatment of automotive exhaust gas. Due to the characteristics of gas diffusion and chamber structure, nitrogen oxide sensors are difficult to achieve high-precision and fast measurement due to strong coupling. In order to solve this problem, this paper proposes a decoupling control strategy for nitrogen oxide sensors. On the one hand, the strategy uses fuzzy feedforward compensation, and on the other hand, it uses PID control based on particle swarm optimization to achieve control. This not only eliminates the coupling between loops, but also improves the control accuracy. The analysis shows that the design of the decoupling control strategy for nitrogen oxide sensors proposed in this paper is feasible, which allows the nitrogen oxide sensors to work in a stable and reliable state.

关键词： Nitrogen-oxygen sensor Coupling Adaptive PID Fuzzy feedforward decoupling

The influencesof different gas flow types on solid oxide fuel cell stack temperature distribution

学校读者我要写书评

暂无评论

The influencesof different gas flow types on solid oxide fue...

第十五届全国氢能会议暨第7届两岸三地氢能研讨会

作者： Huan Cheng Xi Li Jianhua Jiang Yali Liu Jian Li School of automation Key Laboratory of Education Ministry for Image Processing and Intelligent ControlHuazhong University of Science & TechnologyWuhan430074China School of Materials Science and Engineering State Key Laboratory of Material Processing and Die and Mould TechnologyHuazhong University of Science and TechnologyWuhan 430074China

Wavelet-based fusion for optical and MMW radiometric images

学校读者我要写书评

暂无评论

Wavelet-based fusion for optical and MMW radiometric images

International Conference on Microwave and Millimeter Wave Technology Proceedings

作者： Guangfeng Zhang Zuyin Zhang Wei Guo Key Laboratory of State Education Ministry for Image Information Processing and Intelligence Control Department of Electronics and Information Huazhong University of Science and Technology Wuhan Hubei China

ISBN: (纸本)0780384016

On basis of studying the fusion algorithm of multiple sensors, a new approach is presented to fuse the millimeter wave radiometric image and the optical-based simulated image by wavelet transform. The 8mm simulation result shows that fusion image has more complementary information and possesses higher resolution than the measuring image. The fusion image can help to improve the scientific decision-making.

关键词： Microwave radiometry Optical sensors image resolution Millimeter wave technology Millimeter wave measurements Optical imaging Geometrical optics Wavelet transforms intelligent sensors Infrared detectors

Complementary Learning Subnetworks for Parameter-Efficient Class-Incremental Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Depeng Zeng, Zhigang The School of Artificial Intelligence and Automation The Institute of Artificial Intelligence Huazhong University of Science and Technology China The Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Wuhan430074 China

In the scenario of class-incremental learning (CIL), deep neural networks have to adapt their model parameters to non-stationary data distributions, e.g., the emergence of new classes over time. However, CIL models are challenged by the well-known catastrophic forgetting phenomenon. Typical methods such as rehearsal-based ones rely on storing exemplars of old classes to mitigate catastrophic forgetting, which limits real-world applications considering memory resources and privacy issues. In this paper, we propose a novel rehearsal-free CIL approach that learns continually via the synergy between two Complementary Learning Subnetworks. Our approach involves jointly optimizing a plastic CNN feature extractor and an analytical feed-forward classifier. The inaccessibility of historical data is tackled by holistically controlling the parameters of a well-trained model, ensuring that the decision boundary learned fits new classes while retaining recognition of previously learned classes. Specifically, the trainable CNN feature extractor provides task-dependent knowledge separately without interference;and the final classifier integrates task-specific knowledge incrementally for decision-making without forgetting. In each CIL session, it accommodates new tasks by attaching a tiny set of declarative parameters to its backbone, in which only one matrix per task or one vector per class is kept for knowledge retention. Extensive experiments on a variety of task sequences show that our method achieves competitive results against state-of-the-art methods, especially in accuracy gain, memory cost, training efficiency, and task-order robustness. Furthermore, to make the non-growing backbone (i.e., a model with limited network capacity) suffice to train on more incoming tasks, a graceful forgetting implementation on previously learned trivial tasks is empirically investigated. Copyright © 2023, The Authors. All rights reserved.

关键词： Decision making

Re-ID driven localization refinement for person search

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Han, Chuchu Ye, Jiacheng Zhong, Yunshan Tan, Xin Zhang, Chi Gao, Changxin Sang, Nong Key Laboratory of Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Peking University Shanghai Jiao Tong University Megvii Technology

Person search aims at localizing and identifying a query person from a gallery of uncropped scene images. Different from person re-identification (re-ID), its performance also depends on the localization accuracy of a pedestrian detector. The state-of-the-art methods train the detector individually, and the detected bounding boxes may be suboptimal for the following re-ID task. To alleviate this issue, we propose a re-ID driven localization refinement framework for providing the refined detection boxes for person search. Specifically, we develop a differentiable ROI transform layer to effectively transform the bounding boxes from the original images. Thus, the box coordinates can be supervised by the re-ID training other than the original detection task. With this supervision, the detector can generate more reliable bounding boxes, and the downstream re-ID model can produce more discriminative embeddings based on the refined person localizations. Extensive experimental results on the widely used benchmarks demonstrate that our proposed method performs favorably against the state-of-the-art person search methods. Copyright © 2019, The Authors. All rights reserved.

关键词： Machine learning

3D Instances as 1D Kernels

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wu, Yizheng Shi, Min Du, Shuaiyuan Lu, Hao Cao, Zhiguo Zhong, Weicai Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of AIA Huazhong University of Science and Technology China Huawei CBG Consumer Cloud Service Search & Maps BU China

We introduce a 3D instance representation, termed instance kernels, where instances are represented by one-dimensional vectors that encode the semantic, positional, and shape information of 3D instances. We show that instance kernels enable easy mask inference by simply scanning kernels over the entire scenes, avoiding the heavy reliance on proposals or heuristic clustering algorithms in standard 3D instance segmentation pipelines. The idea of instance kernel is inspired by recent success of dynamic convolutions in 2D/3D instance segmentation. However, we find it non-trivial to represent 3D instances due to the disordered and unstructured nature of point cloud data, e.g., poor instance localization can significantly degrade instance representation. To remedy this, we construct a novel 3D instance encoding paradigm. First, potential instance centroids are localized as candidates. Then, a candidate merging scheme is devised to simultaneously aggregate duplicated candidates and collect context around the merged centroids to form the instance kernels. Once instance kernels are available, instance masks can be reconstructed via dynamic convolutions whose weights are conditioned on instance kernels. The whole pipeline is instantiated with a dynamic kernel network (DKNet). Results show that DKNet outperforms the state of the arts on both ScanNetV2 and S3DIS datasets with better instance localization. Code is available: https://***/W1zheng/DKNet. Copyright © 2022, The Authors. All rights reserved.

关键词： Pipelines

A2J-Transformer: Anchor-to-Joint Transformer Network for 3D Interacting Hand Pose Estimation from a Single RGB image

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jiang, Changlong Xiao, Yang Wu, Cunlin Zhang, Mingyang Zheng, Jinghong Cao, Zhiguo Zhou, Joey Tianyi Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China Alibaba Group China Singapore Singapore

3D interacting hand pose estimation from a single RGB image is a challenging task, due to serious self-occlusion and inter-occlusion towards hands, confusing similar appearance patterns between 2 hands, ill-posed joint position mapping from 2D to 3D, etc.. To address these, we propose to extend A2J-the state-of-the-art depth-based 3D single hand pose estimation method-to RGB domain under interacting hand condition. Our key idea is to equip A2J with strong local-global aware ability to well capture interacting hands’ local fine details and global articulated clues among joints jointly. To this end, A2J is evolved under Transformer’s non-local encoding-decoding framework to build A2J-Transformer. It holds 3 main advantages over A2J. First, self-attention across local anchor points is built to make them global spatial context aware to better capture joints’ articulation clues for resisting occlusion. Secondly, each anchor point is regarded as learnable query with adaptive feature learning for facilitating pattern fitting capacity, instead of having the same local representation with the others. Last but not least, anchor point locates in 3D space instead of 2D as in A2J, to leverage 3D pose prediction. Experiments on challenging InterHand 2.6M demonstrate that, A2J-Transformer can achieve state-of-the-art model-free performance (3.38mm MPJPE advancement in 2-hand case) and can also be applied to depth domain with strong generalization. The code is avaliable at https://***/ChanglongJiangGit/A2J-Transformer. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

Model-independent event-based consensus of multiple Euler-Lagrange systems with input disturbances

学校读者我要写书评

暂无评论

Information Sciences 2024年 675卷

作者： Long, Mingkang An, Qing Su, Housheng School of Artificial Intelligence and Automation Image Processing and Intelligent Control Key Laboratory of Education Ministry of China Huazhong University of Science and Technology Wuhan430074 China Artificial Intelligence School Wuchang University of Technology Wuhan430223 China

This paper investigates leaderless consensus (LLC) and leader-follower consensus (LFC) issues of multiple Euler-Lagrange systems (MELSs) with uncertain system parameters and input disturbances. Firstly, by utilizing event-based control strategy, event-based consensus protocols are proposed to achieve LLC and LFC, respectively. Then, to confirm that no Euler-Lagrange (EL) system will exhibit Zeno behavior under the above event-based consensus protocols, we show that the low bound of event-triggering interval is strictly positive for any time instants. Distinct with the existing related works, the proposed event-triggered consensus protocols only rely on local sampled state, and are irrelevant to the system parameters. Finally, some simulation examples are presented to illustrate the feasibility of the proposed event-based consensus algorithms. © 2024 Elsevier Inc.

关键词： Uncertainty analysis

Temporal pattern recognition using spiking neural networks for cortical neuronal spike train decoding

学校读者我要写书评

暂无评论

IFAC Proceedings Volumes 2008年第2期41卷 5203-5208页

作者： Huijuan Fang Yongji Wang Jiping He Shan Liu Key Laboratory for Image Processing and Intelligent Control of Education Ministry of China Department of Control Science and Engineering Huazhong University of Science and Technology Wuhan 430074 China (Tel: 86-27-87557284 The Harrington Department of Bioengineering and the Center for Neural Interface Design of The Biodesign Institute Arizona State University Tempe AZ 85287 USA

Most experimental and decoding algorithm studies of brain neural signals assume that neurons transmit information as a rate coding, but recent studies on the fast cortical computations indicate that temporal coding is probably a more biologically plausible scheme used by neurons. We introduce spiking neural networks (SNN) which consist of spiking neurons propagate information by the timing of spikes to analyze the cortical neural spike trains directly without temporal information lost. The SNN based temporal pattern classification is compared with the conventional artificial neural networks (ANN) based firing rate analysis. The results show that the SNN algorithm can achieve higher accuracy, which demonstrates that temporal coding is a viable code for fast neural information processing and the SNN approach is suitable for recognizing the temporal pattern in the cortical neural signals.

关键词：