检索结果-内蒙古大学图书馆

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

作者： Arjun Teh Wael H. Ali Joshua Rapp Hassan Mansour Carnegie Mellon University Pittsburgh PA USA Mitsubishi Electric Research Laboratories Cambridge MA USA

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

remote temperature sensing of volumetric flows has a variety of applications, such as promoting thermal comfort, heat dissipation, or data center cooling. The emergence of background-oriented schlieren (BOS) imaging in recent years has enabled transparent flow visualization at minor costs. In this paper, we develop a framework for non-invasive volumetric indoor airflow estimation from a single viewpoint using BOS measurements and physics-informed reconstruction. Our framework utilizes a light projector that projects a pattern onto a target back wall and a camera that observes small distortions in the light pattern due to the change in the refractive index of the air as a result of the temperature variation. While the single-view BOS tomography problem is severely ill-posed, we regularize the reconstruction using a physics-informed neural network (PINN) that ensures that the reconstructed airflow is consistent with the coupled Boussinesq approximation of the incompressible Navier– Stokes and the heat transfer equations.

关键词： Temperature measurement Temperature sensors Temperature distribution Volume measurement Neural networks Refractive index Tomography Sensors Speech processing image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Sparrow Mapping using Hyperspectral remote sensing and Deep Learning

Sparrow Mapping using Hyperspectral Remote Sensing and Deep ...

引用

Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA), International conference on

作者： Thilagavathi K Kaavya K Muvin M Harei Sharan S H Department of Electronics and Communication Engineering Kumaraguru College of Technology Coimbatore India

ISBN: (数字)9798331525439

ISBN: (纸本)9798331525446

Biodiversity loss, driven by habitat destruction, poses a significant threat to ecological balance and the survival of endangered species. Accurate identification and monitoring of critical habitats are essential for effective conservation efforts. However, traditional field-based habitat mapping methods are often time-intensive, expensive, and spatially constrained. This work aim to address these limitations by integrating hyperspectral remote sensing with advanced deep-learning methodologies for precise habitat identification and mapping. This work integrates hyperspectral remote sensing with deep learning methodologies, specifically Convolutional Neural Networks (CNNs) and Artificial Neural Networks (ANNs), to analyse high-dimensional hyperspectral data. CNNs are utilized for extracting spatial features, effectively capturing both spectral and spatial variations in the imagery. ANNs are applied for spectral pattern recognition, enabling accurate classification of habitats based on their unique spectral signatures. The pre-processing pipeline includes resizing and grayscale conversion of hyperspectral images to optimize them for deep learning analysis. Validation using field data ensures the reliability of the model outputs. The methodology demonstrated promising results, with ANN achieving a superior accuracy of 99% and CNN achieving an accuracy of 84%. These results are intended to support conservation planning, resource allocation, and evidence-based decision-making. This work contributes to advancing biodiversity conservation and underscores the potential of deep learning in ecological monitoring and management.

关键词： Deep learning Accuracy Biological system modeling Habitats Vegetation mapping Predictive models Feature extraction Convolutional neural networks Monitoring Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

结合特征交互和融合的轻量级变化检测网络

引用

中国图象图形学报 2025年第3期30卷 870-882页

作者：王仁芳杨梓健邱虹王峰高广吴敦浙江万里学院大数据与软件工程学院宁波315100 上海海洋大学信息学院上海201306 宝略科技(浙江)有限公司宁波315042

目的利用深度学习开展变化检测是遥感智能解译热点研究方向之一。针对基于Transformer变化检测模型结构复杂、参数过多、训练耗时的问题,设计了一种融合特征交互和融合的轻量级变化检测网络(feature interaction and fusion lightweight... 详细信息

目的利用深度学习开展变化检测是遥感智能解译热点研究方向之一。针对基于Transformer变化检测模型结构复杂、参数过多、训练耗时的问题,设计了一种融合特征交互和融合的轻量级变化检测网络(feature interaction and fusion lightweight network,FIFLNet)。方法解码器中采用EfficientNet作为特征提取网络,其能利用模型的放缩(model scaling)能力来扩大模型的感受野。然后通过设计通道、像素交互模块(spatial and channel interact block)和浅层跳跃连接(low-level skip-connection)来实现浅层双时相的细节特征交互和上采样阶段的传递,以此增加模型对局部特征的判别精度。此外,利用特征融合分组卷积模块(feature fusion and groups convolution block,FFGCB)对双时相数据进行降维融合,降低了模型计算量。最后,设计了融合上采样模块(fusion upsampling block,FUB)对局部特征与全局特征进行融合还原,同时利用局部特征的细节、纹理来补偿全局特征细节的缺失。结果本文方法在两个遥感影像数据集LEVIR-CD(landearth view image retrieval building change detection dataset)和SYSU-CD(Sun Yat-sen University change detection dataset)上与13种SOTA(state-of-the-art)方法进行比较。实验结果表明,本文方法在变化检测任务中展现出显著优势。在定量评估方面,本文方法在LEVIR-CD和SYSU-CD数据集上的F1值分别达到91.51%和82.19%,较现有最优方法分别提升0.43%和1.58%。同时,模型计算效率优异,仅需1.66 GFLOPs的运算量和0.56 M的参数规模,显著低于所有对比方法。在定性分析方面,本文方法展现出更精确的检测区域定位、更低的漏检率,并能够保留更丰富的细节特征。结论本文提出的轻量级变化检测网络FIFLNet以较少的参数量和每秒浮点运算量获得了优越的性能,改善了小目标漏检、边界误检的情况,能够获得高质量的变化检测结果。

关键词：遥感影像变化检测局部特征特征交互轻量级网络

来源：评论

学校读者我要写书评

暂无评论

Signal processing

引用

Signal processing

作者： Geert Leus

出版社：ELSEVIER

Signal processingincorporates all aspects of the theory and practice of signal processing (analogue and digital). It features original research work, tutorial and review articles, and accounts of practical developments. It is intended for a rapid dissemination of knowledge and experience to engineers and scientists working in the research, development or practical application of signal *** areas covered by the journal include: Signal Theory; Stochastic Processes; Detection and Estimation; Spectral Analysis; Filtering; Signal processing Systems; Software Developments; image processing; pattern recognition; Optical Signal processing; Digital Signal processing; Multi-dimensional Signal processing; Communication Signal processing; Biomedical Signal processing; Geophysical and Astrophysical Signal processing; Earth Resources Signal processing; Acoustic and Vibration Signal processing; Data processing; remote sensing; Signal processing Technology; Speech processing; Radar Signal processing; Sonar Signal processing; Industrial Applications;New Applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Precision Land Use and Land Cover Mapping with a High-Accuracy Time Series LSTM Model

Precision Land Use and Land Cover Mapping with a High-Accura...

引用

IEEE Students' conference on Electrical, Electronics and Computer Science (SCEECS)

作者： Adla Aryan Prakash RV Srinivas Aluvala Venigalla Shashank Chowdary Department of Artificial Intelligence & Machine Learning Ardhaman College of Engineering Hyderabad Telangana India Department of Computer Science & Artificial Intelligence SR University Warangal Telangana India Department of Computer Science and Engineering (Data Science) Rajeev Gandhi Memorial College of Engineering and Technology Nandyal India

ISBN: (数字)9798331529833

ISBN: (纸本)9798331529840

This paper introduces a machine-learning-based LULC classifying model developed with a time series data set, trained on a diverse set of agriculture and non-agricultural classes. It used a strong generalization capability employing an LSTM architecture; that is, it achieved accuracy of 90.04% in training and also validation accuracy of 90.55%. The ultimate accuracy was obtained at the minimal loss of 0.1909 at the accuracy of 95.58%, thereby validating its usability for LULC pattern classification. This model has a wide-range of potential application in environmental monitoring and resource management. Future work involves introducing more variables into the dataset, the inclusion of other environmental factors, and fine-tuning of the model towards more general applications in geospatial applications. Thus, this work advances the current state of remote sensing and LULC classification to present an effective tool in sustainable land management.

关键词： Deep learning Accuracy image processing Time series analysis Land surface Spatial databases Environmental monitoring Resource management remote sensing Long short term memory

来源：评论

学校读者我要写书评

暂无评论

基于光学遥感图像的土地利用空间格局识别

引用

勘察科学技术 2025年第1期 68-74页

作者：李彦粉北京威特空间科技有限公司北京市102600

土地利用空间在格局上的分界差异很小,当前土地利用空间格局识别过程中主要使用以电磁波为主的普通遥感图像,受困于采集的土地空间遥感特征尺度差异大,几何偏差突变区域多,导致土地利用空间格局识别效果差,该文提出了基于光学遥感图像... 详细信息

土地利用空间在格局上的分界差异很小,当前土地利用空间格局识别过程中主要使用以电磁波为主的普通遥感图像,受困于采集的土地空间遥感特征尺度差异大,几何偏差突变区域多,导致土地利用空间格局识别效果差,该文提出了基于光学遥感图像的土地利用空间格局识别方法。该方法利用无人机采集遥感图像,设计一种根据基准图像和校正图像特征对应关系的图像拼接技术,以克服当前大差异土地光学遥感图像几何校正的弊端;设计了一种暗通道和双边滤波相结合的方法,对校正后的图像进行增强处理,该方法在滤波过程中考虑了像素之间的空间距离和灰度相似性,双边滤波能够降低图像噪声并保持图像细节,同时避免了传统高斯滤波对边缘信息的模糊处理;通过融合多尺度与主方向因子的灰度共生矩阵纹理特征,完成土地利用空间格局的识别。实验结果表明,该方法对土地利用空间格局识别准确率高、效率高且效果好。

关键词：光学遥感图像土地利用空间格局识别几何校正图像增强

来源：评论

学校读者我要写书评

暂无评论

Implicit neural representation model for camera relocalization in multiple scenes

引用

pattern recognition 2025年 168卷

作者： Yao, Shun Cheng, Yongmei Yang, Fei Mozerov, Mikhail G. School of Automation NWPU Shaanxi Xi'an710129 China Computer Vision Center UAB Barcelona08193 Spain College of Computer Science Nankai University Tianjin China NKIARI Shenzhen Futian China Institute for Information Transmission Problems Russian Academy of Sciences Moscow Russia

One critical task in remote sensing is determining the position of a video camera relative to the scene depicted in a series of images captured by the camera. Classical approaches often necessitate pre-built scenario representations and the implementation of complex, time-consuming algorithms. Recent methods utilizing scene coordinate (SC) regression-based models have demonstrated promising performance in visual relocalization regarding accuracy and efficiency for a single scene. However, extending SC regression models to multiple scenes typically requires retraining model parameters or constructing reference landmarks, which is a time-consuming process. This paper proposes representing multiple scenes within a global reference coordinate system to efficiently train a single SC regression model in one training procedure. We encode scene information in scene embeddings as a prior condition for our model predictions. We design a scene-conditional regression-adjust (SCRA) module to adapt the model to the scene embedding by dynamically generating parameters during inference. Additionally, we employ modulation and complement modules to enhance the model's prediction applicability at both the image sample and scene levels. The modulation module adjusts the amplitude, phase, and frequency of the data flow for each input image, while the complement module derives scene-specific coordinate biases to reduce distribution differences between scenes. Extensive experiments on indoor and outdoor datasets validate our model's efficiency and accuracy in multi-scene visual relocalization. Compared to the state-of-the-art MS-Transformer model, our model requires less training time and achieves more accurate relocalization results, with reductions in average median errors of position and rotation by 50.0% and 52.0% on the Cambridge Landmarks dataset, and by 61.1% and 73.9% on the 7Scenes dataset. Compared to the separately trained advanced FeatLoc++Au model, our model achieves relative

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

Quantum Machine Learning in Crop Disease Monitoring: Opportunities and Challenges to Practical Implementation

Quantum Machine Learning in Crop Disease Monitoring: Opportu...

引用

International conference on Computational Cybernetics (ICCC)

作者： Yue Wu Attila Nagy Zoltan Rajnai Beatrix Fregan Katalin Takács-György Bánki Donát Faculty of Mechanical and Safety Engineering Obuda University Budapest Hungary Doctoral School on Safety and Security Science Obuda University Budapest Hungary University of Public Service Budapest Hungary Keleti Faculty of Business and Management Obuda University Budapest Hungary

ISBN: (数字)9798331502461

ISBN: (纸本)9798331502478

Agriculture is increasingly threatened by diverse global security challenges, including climate change, biodiversity loss, and health security risks, underscoring the need for sustainable practices to ensure food security for future generations. Advanced technologies such as quantum machine learning (QML) offer transformative potential in addressing these challenges by enhancing crop disease monitoring systems. This paper explores the implementation of QML in agriculture, with a particular focus on its application to crop disease detection and prevention. Leveraging image processing, sensory data analysis, and spectral imaging, QML algorithms demonstrate superior accuracy and efficiency in early disease recognition, surpassing traditional methods. A case study of Hungary's crop disease monitoring system illustrates the practical benefits of integrating QML with technologies such as remote sensing and IoT. By enabling real-time monitoring and predictive analytics, QML not only aids in mitigating crop losses but also contributes to sustainable agricultural practices and global food security. Despite current limitations in quantum hardware, advancements in QML present promising opportunities for revolutionizing agricultural systems and ensuring resilience against evolving threats.

关键词： Quantum computing Crops Food security Machine learning Agriculture Sustainable development Monitoring remote sensing Diseases Resilience

来源：评论

学校读者我要写书评

暂无评论

Methods for Updating Forest Inventory Data Through Multi-temporal Sentinel-2 image Analysis 15th

Methods for Updating Forest Inventory Data Through Multi-tem...

引用

15th International conference on Soft Computing and pattern recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Bykova, Daria Denisova, Anna Fedoseev, Victor Korchikov, Evgeny Samara National Research University Moskovskoe Shosse 34 Samara 443086 Russian Federation

ISBN: (纸本)9783031810855

Paper presents methods for utilizing the multi-temporal imagery from the Sentinel-2 satellite to detect changes in forest cover, facilitating the updating of forest inventory data. Sentinel-2 images from various dates are combined into a single composite image that is further used as a feature vector set. Digital forest inventory data, which may not always correspond to the actual ground state due to potential inaccuracies or infrequent observations, is employed to identify monospecies regions and assess the classifier’s performance. Misclassified areas within the forest could signal changes in the forest ecosystem. Experimental methods have been developed to filter these areas, enhancing the accuracy in discerning actual changes from classifier errors within complex mixed forest regions. The empirical data presented within this study serves to validate the effectiveness of the proposed solutions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Change Detection Forest Inventory Data remote sensing Sentinel-2 Spatial Classification Temporary Analysis

来源：评论

学校读者我要写书评

暂无评论

VQARS: Visual Question Answering for remote sensing as a Domain of Choice Incorporating Semantic Intelligence 15th

VQARS: Visual Question Answering for Remote Sensing as a Dom...

引用

15th International conference on Soft Computing and pattern recognition, SoCPaR 2023 and 14th World Congress on Nature and Biologically Inspired Computing, NaBIC 2023

作者： Manoj Kumar, H.S. Deepak, Gerard Santhanavijayan, A. Department of Computer Science and Engineering PES University Bangalore India Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Computer Science and Engineering National Institute of Technology Tiruchirappalli Tiruchirappalli India

ISBN: (纸本)9783031889912

The need for a visual question answering framework in the era of Web 3.0 is realized by using staged addition of auxiliary knowledge in the perspective of two distinct datasets namely the dataset of documents and the images which is enriched through informative term extraction and further using structural topic modelling, Wikidata, NELL and Linked Open Data (LOD) Cloud, knowledge stores and tag generation to classify the dataset obtained informative terms. Subsequently, the image dataset terms are subjected to extraction of categories and labels to contribute to the informative terms. The random forest classifier at either ends of the two datasets helps in controlled classification of the datasets independently and assures computational inexpensiveness because of the lightweighted nature of the machine learning classifiers. The RNN to classify the tags helps in automatizing the exponentially generated tags and semantic network formulated helps in organizing the auxiliary knowledge. Semantic relevance computation using Jaccard Similarity, Morisitas Overlap Index, NPMI and SOC-PMI facilitates in semantic reasoning. The proposed VQARS model's achieves an overall precision of 94.81%;recall of 95.75%;average accuracy of 95.28%;F-measure of 95.28%;and the lowest FDR of 0.06. In comparison to the other baseline models, this model provides the best-in-class outcomes. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Knowledge-centric model remote sensing semantic search visual question answering Web 3.0

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：