检索结果-内蒙古大学图书馆

CodeEnhance: A Codebook-Driven Approach for Low-Light Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Xu Hou, XianXu Lai, Zhihui Zhou, Jie Zhang, Ya-Nan Pedrycz, Witold Shen, Linlin The Computer Vision Institute College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518060 China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Shenzhen518060 China The Department of Electrical & Computer Engineering University of Alberta University of Alberta Canada

Low-light image enhancement (LLIE) aims to improve low-illumination images. However, existing methods face two challenges: (1) uncertainty in restoration from diverse brightness degradations;(2) loss of texture and color information caused by noise suppression and light enhancement. In this paper, we propose a novel enhancement approach, CodeEnhance, by leveraging quantized priors and image refinement to address these challenges. In particular, we reframe LLIE as learning an image-to-code mapping from low-light images to discrete codebook, which has been learned from high-quality images. To enhance this process, a Semantic Embedding Module (SEM) is introduced to integrate semantic information with low-level features, and a Codebook Shift (CS) mechanism, designed to adapt the pre-learned codebook to better suit the distinct characteristics of our low-light dataset. Additionally, we present an Interactive Feature Transformation (IFT) module to refine texture and color information during image reconstruction, allowing for interactive enhancement based on user preferences. Extensive experiments on both real-world and synthetic benchmarks demonstrate that the incorporation of prior knowledge and controllable information transfer significantly enhances LLIE performance in terms of quality and fidelity. The proposed CodeEnhance exhibits superior robustness to various degradations, including uneven illumination, noise, and color distortion. Copyright © 2024, The Authors. All rights reserved.

关键词： Image enhancement

Attention mechanism meets with hybrid dense network for Hyperspectral Image Classification

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ahmad, Muhammad Mehmood Khan, Adil Mazzara, Manuel Distefano, Salvatore Roy, Swalpa Kumar Wu, Xin Department of Computer Science National University of Computer and Emerging Sciences Islamabad Chiniot-Faisalabad Campus Chiniot35400 Pakistan Institute of Data Science and Artificial Intelligence Innopolis University Innopolis420500 Russia Institute of Software Development and Engineering Innopolis University Innopolis420500 Russia Dipartimento di Matematica e Informatica MIFT University of Messina Messina98121 Italy Department of Computer Science and Engineering Jalpaiguri Government Engineering College West Bengal 735102 India School of Information and Electronics Beijing Institute of Technology Beijing100081 China

The nonlinear relation between the spectral information and the corresponding objects (complex physiognomies) makes pixel-wise classification challenging for conventional methods. To deal with nonlinearity issues in Hyperspectral Image Classification (HISC), Convolutional Neural Networks (CNN) are more suitable, indeed. However, fixed kernel sizes make traditional CNN too specific, neither flexible nor conducive to feature learning, thus impacting on the classification accuracy. The convolution of different kernel size networks may overcome this problem by capturing more discriminating and relevant information. In light of this, the proposed solution aims at combining the core idea of 3D and 2D Inception net with the Attention mechanism to boost the HSIC CNN performance in a hybrid scenario. The resulting attention-fused hybrid network (AfNet) is based on three attention-fused parallel hybrid subnets with different kernels in each block repeatedly using highlevel features to enhance the final ground-truth maps. In short, AfNet is able to selectively filter out the discriminative features critical for classification. Several tests on HSI datasets provided competitive results for AfNet compared to state-of-the-art models. The proposed pipeline achieved, indeed, an overall accuracy of 97% for the Indian Pines, 100% for Botswana, 99% for Pavia University, Pavia Center, and Salinas datasets. © 2022, CC BY.

关键词： Convolution

MDE for Machine Learning-Enabled software Systems: A Case Study and Comparison of MontiAnna & ML-Quadrat

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Kirchhof, Jörg Christian Moin, Armin Badii, Atta Günnemann, Stephan Challenger, Moharram Evgeny Kusmenko Jonas Ritz Bernhard Rumpe Software Engineering RWTH Aachen University Germany School of Computation Information and Technology Technical University of Munich University of Antwerp & Flanders Make Germany Belgium Department of Computer Science University of Reading United Kingdom School of Computation Information and Technology Technical University of Munich Munich Data Science Institute Germany Department of Computer Science University of Antwerp Flanders Make Belgium

In this paper, we propose to adopt the MDE paradigm for the development of Machine Learning (ML)-enabled software systems with a focus on the Internet of Things (IoT) domain. We illustrate how two state-of-the-art open-source modeling tools, namely MontiAnna and ML-Quadrat can be used for this purpose as demonstrated through a case study. The case study illustrates using ML, in particular deep Artificial Neural Networks (ANNs), for automated image recognition of handwritten digits using the MNIST reference dataset, and integrating the machine learning components into an IoT-system. Subsequently, we conduct a functional comparison of the two frameworks, setting out an analysis base to include a broad range of design considerations, such as the problem domain, methods for the ML integration into larger systems, and supported ML methods, as well as topics of recent intense interest to the ML community, such as AutoML and MLOps. Accordingly, this paper is focused on elucidating the potential of the MDE approach in the ML domain. This supports the ML-engineer in developing the (ML/software) model rather than implementing the code, and additionally enforces reusability and modularity of the design through enabling the out-of-the-box integration of ML functionality as a component of the IoT or cyber-physical systems. © 2022, CC BY-NC-SA.

关键词： Internet of things

Time-Aware Missing Traffic Flow Prediction for Sensors with Privacy-Preservation

学校读者我要写书评

暂无评论

Time-Aware Missing Traffic Flow Prediction for Sensors with ...

作者： Lianyong Qi Fan Wang Xiaolong Xu Wanchun Dou Xuyun Zhang Mohammad R.Khosravi Xiaokang Zhou School of Computer Science Qufu Normal University School of Computer and Software Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology Nanjing University of Information Science and Technology State Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Department of Computing Macquarie University Department of Computer Engineering Persian Gulf University Department of Electrical and Electronic Engineering Shiraz University of Technology Faculty of Data Science Shiga University

With the continuous development of Io T, a number of sensors establish on the roadside to monitor traffic conditions in real time. The continuously traffic data generated by these sensors makes traffic management feasible. However, loss of data may occur due to inevitable sensor failure, impeding traffic managers to understand traffic dynamics clearly. In this situation, it is becoming a necessity to predict missing traffic flow accurately for effective traffic management. Furthermore, the traffic sensor data are often distributed and stored by different agencies,which inhibits the multi-party sensor data sharing significantly due to privacy concerns. Therefore, it has become a major obstacle to balance the tradeoff between data sharing and vehicle privacy. In light of these challenges, we propose a privacyaware and accurate missing traffic flow prediction approach based on time-aware Locality-Sensitive Hashing technique. At last, we deploy a set of experiments based on a real traffic dataset. Experimental reports demonstrate the feasibility of our proposal in terms of traffic flow prediction accuracy and efficiency while guaranteeing sensor data privacy.

关键词： Traffic flow prediction Multi-party sensors Privacy Locality-Sensitive Hashing

NuSegDG: Integration of Heterogeneous Space and Gaussian Kernel for Domain-Generalized Nuclei Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lou, Zhenye Xu, Qing Jiang, Zekun He, Xiangjian Chen, Zhen Wang, Yi Li, Chenxin He, Maggie M. Duan, Wenting Sichuan University Pittsburgh Institute Sichuan University Chengdu China School of Computer Science University of Nottingham Ningbo China Zhejiang Ningbo China West China Biomedical Big Data Center West China Hospital Sichuan University Chengdu China Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong School of Software Dalian University of Technology Dalian116600 China Department of Electronic Engineering The Chinese University of Hong Kong 999077 Hong Kong Department of Cardiology Gold Coast University Hospital QLD Australia School of Computer Science University of Lincoln LincolnLN6 7TS United Kingdom

Domain-generalized nuclei segmentation refers to the generalizability of models to unseen domains based on knowledge learned from source domains and is challenged by various image conditions, cell types, and stain strategies. Recently, the Segment Anything Model (SAM) has made great success in universal image segmentation by interactive prompt modes (e.g., point and box). Despite its strengths, the original SAM presents limited adaptation to medical images. Moreover, SAM requires providing manual bounding box prompts for each object to produce satisfactory segmentation masks, so it is laborious in nuclei segmentation scenarios. To address these limitations, we propose a domain-generalizable framework for nuclei image segmentation, abbreviated to NuSegDG. Specifically, we first devise a Heterogeneous Space Adapter (HS-Adapter) to learn multi-dimensional feature representations of different nuclei domains by injecting a small number of trainable parameters into the image encoder of SAM. To alleviate the labor-intensive requirement of manual prompts, we introduce a Gaussian-Kernel Prompt Encoder (GKP-Encoder) to generate density maps driven by a single point, which guides segmentation predictions by mixing position prompts and semantic prompts. Furthermore, we present a Two-Stage Mask Decoder (TSM-Decoder) to effectively convert semantic masks to instance maps without the manual demand for morphological shape refinement. Based on our experimental evaluations, the proposed NuSegDG demonstrates state-of-the-art performance in nuclei instance segmentation, exhibiting superior domain generalization capabilities. The source code is available at https://***/xq141839/NuSegDG. © 2024, CC BY.

关键词： Decoding

Correction: DeepDet: YAMNet with BottleNeck Attention Module (BAM) for TTS synthesis detection

学校读者我要写书评

暂无评论

EURASIP Journal on Audio, Speech, and Music Processing 2024年第1期2024卷 1-1页

作者： Mahum, Rabbia Irtaza, Aun Javed, Ali Mahmoud, Haitham A. Hassan, Haseeb Computer Science Department UET Taxila Taxila Pakistan Software Engineering Department UET Taxila Taxila Pakistan Industrial Engineering Department College of Engineering King Saud University Riyadh Saudi Arabia College of Big Data and Internet Shenzhen Technology University (SZTU) Shenzhen China

Correction: Uncertainty Modelling in Performability Prediction for Safety-Critical Systems

学校读者我要写书评

暂无评论

Arabian Journal for science and engineering 2024年第2期50卷 1353-1353页

作者： Ahamad, Shakeel Gupta, Ratneshwer Department of Data Science Noida Institute of Engineering and Technology Greater Noida India Software Quality Assurance Lab School of Computer and Systems Sciences Jawaharlal Nehru University New Delhi India

Classification of Awareness in Disorders of Consciousness Using Common Spatial Pattern

学校读者我要写书评

暂无评论

Classification of Awareness in Disorders of Consciousness Us...

International Winter Workshop on Brain-Computer Interface (BCI)

作者： Chaewon Lee Sunho Lee Jitka Annen Ji-Hoon Jeong Marcello Massimini Silvia Casarotto Melanie Boly Olivier Bodart Aurore Thibaut Mario Rosanova Steven Laureys Olivia Gosseries Minji Lee Dept. Data Science The Catholic University of Korea Bucheon Republic of Korea Dept. Biomedical Software Engineering The Catholic University of Korea Bucheon Republic of Korea Dept. GIGA-Research & Neurology University and University Hospital of Liege Liege Belgium Dept. Data Analysis University of Ghent Ghent Belgium Dept. Computer Science Chungbuk National University Cheongju Republic of Korea Dept. Biomedical and Clinical Sciences “L. Sacco” University of Milan Milan Italy Dept. Psychiatry Wisconsin Institute for Sleep and Consciousness University of Wisconsin Madison USA Coma Science Group GIGA-Consciousness GIGA Research Center University of Liège Liège Belgium Centre du Cerveau Neurology department University Hospital of Liège Liège Belgium

ISBN: (数字)9798331521929

ISBN: (纸本)9798331521936

Advances in intensive care have improved the survival rate of patients with severe acute brain injury, but diagnostic errors for patients with disorders of consciousness are still high. Accurate diagnosis of these patients is very important because effective treatment can vary depending on the diagnosis. In this study, we propose a framework for classifying unresponsive wakefulness syndrome and minimally conscious state, focusing on awareness. In particular, power spectral density and common spatial patterns were used together, considering that spatial information is a key feature in consciousness. The 16 patients with unresponsive wakefulness syndrome and 14 with minimally conscious state underwent resting-state electroencephalography measurements. In addition, we compared the performance by utilizing each frequency (delta, theta, alpha, beta, gamma bands) related to consciousness. As a result, the highest accuracy of 95.06% was achieved by the EEGNet classifier, especially in the beta frequency band. We demonstrated that spatial information is very important in consciousness, as we observed that classification performance improved when common spatial patterns were used. These results provide insight into various frameworks for diagnosing patients with disorders of consciousness and may help patients survive by increasing the diagnosis rate in the future.

关键词： Deep learning Accuracy Biological system modeling Pattern classification Focusing Signal processing Brain modeling Particle measurements Electroencephalography Robustness

Vacancy-Type Defects in HfO2 Layers and Their Role in Amorphous-to-Crystalline Transition Studied by Monoenergetic Positron Beams

学校读者我要写书评

暂无评论

IEEE Transactions on Semiconductor Manufacturing 2025年

作者： Uedono, Akira Kawakami, Kazuya Kuribara, Tatsunori Hasunuma, Ryu Harashima, Yosuke Shigeta, Yasuteru Ni, Zeyuan Yan, Ruolin Nakamura, Hideo Notake, Akira Moriya, Tsuyoshi Michishio, Koji Ishibashi, Shoji University of Tsukuba Faculty of Pure and Applied Science Tsukuba Ibaraki305-8573 Japan Nara Institute of Science and Technology Graduate School of Science and Technology Ikoma Nara630-0192 Japan Tokyo Electron Technology Solutions Ltd S-Technology Development Center Nirasaki Yamanashi650 Japan Tokyo Electron Technology Solutions Ltd Development Department1 Nirasaki Yamanashi407-0192 Japan Tokyo Electron Ltd Digital Design Center Sapporo Hokkaido060-0003 Japan Tokyo Electron Technology Solutions Ltd Software and Data Engineering Division Nirasaki Yamanashi407-0192 Japan National Institute of Advanced Industrial Science and Technology Research Institute for Measurement and Analytical Instrumentation Tsukuba Ibaraki305-8568 Japan National Institute of Advanced Industrial Science and Technology Research Center for Computational Design of Advanced Functional Materials Tsukuba Ibaraki305-8568 Japan

The annealing properties of vacancies in HfO2 layers deposited on Si substrates and their role in amorphous-to-crystalline transition were studied with monoenergetic positron beams. HfO2 layers with a thickness of 4.30 nm were fabricated by the atomic layer deposition technique. The major vacancytype defects in these layers were identified as a Hf vacancy (VHf) coupled with multiple oxygen vacancies (VOs) and larger vacancy clusters. After annealing at 500°C, the concentration of vacancy clusters started to increase, which was attributed to the agglomeration of intrinsic open spaces in the amorphous phase upon the phase transition from amorphous to monoclinic crystalline phases. The amorphous-crystalline transition started near the interface between the HfO2 layer and bottom electrodes (TiN). After the crystallization, the concentration of vacancy clusters decreased as the annealing temperature increased. © 2025 IEEE.

关键词： Hafnium oxides