检索结果-内蒙古大学图书馆

24th ISPRS Congress on Imaging Today, Foreseeing Tomorrow

作者： Menna, F. Nocerino, E. Malek, S. Remondino, F. Schiaparelli, S. Bruno Kessler Fdn FBK 3D Opt Metrol 3DOM Unit Trento Italy Univ Sassari Dipartimento Sci Umanist & Sociali Sassari Italy Univ Genoa DISTAV Genoa Italy Italian Natl Antarctic Museum MNA Sect Genoa Genoa Italy

Long-term monitoring projects are becoming more than ever crucial in assessing the effects of climate change on marine communities, especially in Antarctica, where these changes are expected to be particularly dramatic. Detailed studies of the Antarctic benthos are in fact particularly important for a better understanding of benthos dynamics and potential climate-driven shifts. Here, due to the extreme fragility of benthic communities, non-destructive techniques represent the best solution in long-term monitoring programs. In this paper we report new results from 2017, 2018, 2019 photogrammetric campaigns within the Italian National Antarctic Research Program (PNRA). A new protocol of data acquisition and multi-temporal processing that provides co-registered 3D point clouds between the three years without control points nor direct georeferencing methods is presented. This is achieved by adding a level of image understanding leveraging semantic segmentation with convolutional neural network (CNN) of the benthic features. Slow growing (estimated less than a mm per year) organisms, such as Corallinales (Rhodophyta algae), represent a natural stable pattern, leveraged to automatically orient in the same reference system the photogrammetric surveys of different epochs. This approach is also proved to be effective in improving the orientation of adjacent strips acquired within the same campaign. Within the paper an in depth analysis of the achieved results shows the effectiveness of the implemented procedure.

关键词： Underwater photogrammetry change detection long-term monitoring antarctica image segmentation CNN

来源：评论

学校读者我要写书评

暂无评论

Target detection of remote sensing images based on deep learning method and system 3

Target detection of remote sensing images based on deep lear...

引用

3rd International Conference on Advanced Information Science and System, AISS 2021

作者： Wang, Su Jun Ping, Yi Fan Chen, Gang Yang, Li Wen, Wei Xu, Chang Zhi Shao, Ying Zhao China Academy of Space Technology China

ISBN: (纸本)9781450385862

With the rapid growth of remote sensing image data, it is very important to find a way to extract and recognize the target quickly and accurately from the massive remote sensing data. In recent years, the development of deep learning has provided an effective way for target detection of remote sensing images. This paper applies deep learning technology to target detection of remote sensing images, and constructs a target detection system software which integrates sample labeling, data set construction, pretreatment of training sample, training algorithm, migration learning, target recognition and post processing. It provides technical support to the tasks of classification, information extraction and change detection of remote sensing image. The experimental results show that the target recognition system of remote sensing images has high precision in the scene classification and specific target detection of high-resolution remote sensing images. © 2021 ACM.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Accurate Angle Detection of Rotated Rectangles Using Hybrid Architecture of Convolutional Neural Networks with Multi-Layer Perceptron and SVR

Accurate Angle Detection of Rotated Rectangles Using Hybrid ...

引用

International Conference on Emerging Trends in Networks and Computer Communications (ETNCC)

作者： Nayeem Ahmed Faculty of Electrical and Computer Engineering Technische Universität Dresden Dresden Germany

ISBN: (数字)9798350353266

ISBN: (纸本)9798350353273

This study presents a novel approach for detecting the angles of the rotated rectangles precisely using the hybrid architecture of Convolutional Neural Networks (CNN) with Multi-Layer Perceptron (MLP) and Support Vector Regression (SVR). This work also shows the comparative assessment between the two hybrid models, CNN & MLP only and CNN & MLP along with the SVR for unrolling the angles of the rectangles. In the automated image analysis and pattern recognition domain, the complexity of rotated rectangles-especially in different orientations and scales-presents formidable challenges. Our study begins with the dataset comprised of 10000 images of rectangles with varying rotation angles and coordinates. Then, CNN, an effective model in the image analysis and computer vision field, effectively captures the spatial dependencies and characteristics of rotated rectangles from the raw images by extracting and learning the hierarchical feature representations. To further process the pieces of information, the MLP and SVR are used, giving the learning model more depth and improving its capacity to recognize complex patterns. Evaluation metrics such as MSE, RMSE, MAPE, MAE, and $\mathbf{R}^pattern$ determine the model's accuracy. This research enhances the fields of machine learning and image processing and also potentially benefits robotics, com-puter vision, and remote sensing-all of which depend on precise geometric interpretation. The evaluation metrics corroborate that the algorithm based on CNN and MLP along with the SVR has better accuracy compared to the model that relies only on CNN and MLP.

关键词： Measurement Analytical models image analysis Accuracy Computational modeling Computer architecture Feature extraction Data models pattern recognition Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

remote sensing image Captioning Using Hire-MLP

Remote Sensing Image Captioning Using Hire-MLP

引用

Computer Vision, image and Deep Learning (CVIDL), International Conference on

作者： Yichao Wei Lin Li Shengling Geng School of Computer Science Qinghai Normal University Xining China The State Key Laboratory of Tibetan Intelligent Information Processing and Application Xining China Academy of Plateau Science and Sustainability Qinghai Normal University Xining China National Qinghai-Tibet Plateau Scientific Data Center Qinghai Sub-center Xining China

remote sensing image captioning has been widely applied to traffic management, geographic research, etc. Although the neural network approach has been successfully improving the performance of the remote sensing image captioning system, remote sensing image captioning is still facing object identification challenges due to the small size of objects, uneven distribution, and high coupling with the surrounding image background. In this paper, we propose a novel remote sensing image captioning encoder-decoder model Hierarchical rearrangement-Multi-Layer Perceptron (HMLP) whose encoder adapts hierarchical rearrangement-multi-layer perceptron to improve the capability of objection recognition. Extensive experiments have been conducted to testify HMLP by three datasets RSCID, UCM-caption, and NWPU-caption. Results show that HMLP outperforms many image captioning systems in the evaluation metrics BLEU4, METEOR, ROUGE-L, and CIDEr.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Track Hyperspectral Videos via Segment Anything Model 2

Track Hyperspectral Videos via Segment Anything Model 2

引用

Workshop on Hyperspectral image and Signal processing: Evolution in remote sensing, WHISPERS

作者： Zixuan Qian Zongming Li Bin Feng Wenyu Liu Xinggang Wang Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798331513139

ISBN: (纸本)9798331513146

Hyperspectral object tracking aims to locate objects through spectral and spatial information, especially those that share similar appearances. However, due to the scarcity of hyperspectral image datasets, existing tracking methods are usually trained on top of RGB trackers. In this paper, we propose HyperSAM to demonstrate that, with appropriate false color conversion, the problem can be effectively addressed by the state-of-the-art video segmentation model without further training. We utilize SAM2 framework and leverage the power of its memory bank, selecting relevant past frames in a Siamese way to consolidate new memories. This selection strategy shows effectiveness in solving distraction and occlusion problems in tracking. The aspiration of this paper is to serve as a benchmark in achieving high-precision hyperspectral object tracking. The performance of the framework is evaluated through its application to a challenging hyperspectral video dataset containing several challenging factors. Quantitative experiments are conducted on different bands and demonstrate that the proposed method achieves competitive performances compared with other state-of-the-art trackers.

关键词： Training image segmentation Target tracking image color analysis Target recognition Conferences Signal processing Benchmark testing Object tracking Hyperspectral imaging

来源：评论

学校读者我要写书评

暂无评论

Evolutionary image Registration: A Review

引用

SENSORS 2023年第2期23卷 967-967页

作者： Cocianu, Catalina-Lucia Uscatu, Cristian Razvan Stan, Alexandru Daniel Bucharest Univ Econ Studies Dept Econ Informat & Cybernet Bucharest 010552 Romania

image registration is one of the most important image processing tools enabling recognition, classification, detection and other analysis tasks. Registration methods are used to solve a large variety of real-world problems, including remote sensing, computer vision, geophysics, medical image analysis, surveillance, and so on. In the last few years, nature-inspired algorithms and metaheuristics have been successfully used to address the image registration problem, becoming a solid alternative for direct optimization methods. The aim of this paper is to investigate and summarize a series of state-of-the-art works reporting evolutionary-based registration methods. The papers were selected using the PRISMA 2020 method. The reported algorithms are reviewed and compared in terms of evolutionary components, fitness function, image similarity measures and algorithm accuracy indexes used in the alignment process.

关键词： evolutionary algorithms image registration fitness functions image similarity indicators accuracy index

来源：评论

学校读者我要写书评

暂无评论

Federated Target recognition for Multiradar Sensor Data Security

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND remote sensing 2023年 61卷 1页

作者： Song, Yafei Dong, Ganggang Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China

A single sensor radar can no longer satisfy the increasingly complex electromagnetic environment. More attention is paid to radar sensor networks, which can obtain more information from different nodes and points of view. Moreover, better detection and tracking performance can be achieved through resource sharing and complementary advantages (joint learning). How to improve the utilization efficiency of multiple radar sensors with limited resources is an open problem, which can be transformed into joint learning in scenarios, such as limited training data or imbalanced samples. This article presents a distributed learning model to solve these problems. It has three phases. Self-reweighting (SR) loss is developed to dynamically rebalance the gradients of positive and negative samples for each category, after which the imbalance of samples can be alleviated. An image generation technique via target reimaging addresses the problem of limited samples. SR loss and image generation are then unified in a federated learning (FL) framework. The classifier is adjusted using virtual representations to further improve learning efficiency. Comparative studies on the Moving and Stationary Target Acquisition and recognition (MSTAR) dataset demonstrate the advantages of the proposed method.

关键词： Radar Radar imaging Training Servers image reconstruction Federated learning Target recognition Deep learning federated algorithm finite samples label imbalance radar networking

来源：评论

学校读者我要写书评

暂无评论

Global and Local Contrastive Learning for Classification and Segmentation of Mixed-type Wafer Bin Map Defect patterns

Global and Local Contrastive Learning for Classification and...

引用

IEEE International Conference on Industrial Engineering and Engineering Management

作者： Shantong Yin Rui Wang Peng Li Department of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen China

ISBN: (数字)9798350386097

ISBN: (纸本)9798350386103

Defect pattern recognition in semiconductor wafer bin maps (WBMs) presents a formidable challenge in the integrated circuit manufacturing industry. Precise wafer defect pattern classification and segmentation can trace the root cause of defect patterns in the manufacturing process, thereby mitigating cost losses and augmenting efficiency and quality of products. When different defects are mixed on the same wafer, the WBM becomes increasingly intricate, which further increases the difficulty of recognition. Existing supervised learning methods require a large number of labeled samples, which is undoubtedly labor-intensive. In this paper, we propose a self-supervised contrastive learning framework to classify and segment different mixed-type WBM defect patterns by combining global and local contrastive learning modules. Specifically, global contrastive learning module is designed to learn image-level representation, while local contrastive learning module is used to better understand the structure of local regions, which contributes to image segmentation task. Experimental results show that our model works well only requiring little labeled samples and abundant unlabeled samples.

关键词： Semiconductor device modeling image segmentation Computational modeling Supervised learning pattern classification Contrastive learning Data models pattern recognition remote sensing Optimization

来源：评论

学校读者我要写书评

暂无评论

Classification of hyperspectral remote sensing image via rotation-invariant local binary pattern-based weighted generalized closest neighbor

引用

JOURNAL OF SUPERCOMPUTING 2021年第6期77卷 5528-5561页

作者： Sharma, Monika Biswas, Mantosh Natl Inst Technol Kurukshetra Dept Comp Engn Kurukshetra Haryana India

In this article, the authors suggested a rotation-invariant local binary pattern-based weighted generalized closest neighbor (RILBP-WGCN) method for HSI classification. The proposed RILBP is an enhanced texture-based classification paradigm that utilizes local binary pattern filter for some designated bands to generate a broad sketch of spatial texture information. Likewise, the proposed WGCN technique efficiently maintains the spatial uniformity between the nearby pixels via utilizing a local weight scheme and point-to-set distance. Also, as a postprocessing step, a label enhancement method is included for additional enhancement of the label uniformity as well as increases the performance of classification method. The color composite remotely sensed image of the initial three subsequent bands is segmented into several consistent regions by utilizing the graph-based superpixel segmentation technique. Then, extracted super pixels have been made extra homogeneous by utilizing a segment grouping process. Finally, advanced decision-level fusion is also applied on the retrieved local LBP features and unique spectral features, where linear opinion pool executes a serious role for concatenating the probabilistic outcomes of numerous spectral as well as texture features. The authors evaluated the proposed technique by comparing them with the seven competing methods on numerous datasets related to HSI classification. Evaluation results confirmed that the classification effects of proposed RILBP-WGCN algorithm are significantly better in contrast to other competing classification schemes.

关键词： Hyperspectral image (HSI) classification Superpixel Local binary pattern (LBP) Rotation invariance Feature extraction pattern recognition

来源：评论

学校读者我要写书评

暂无评论

GeoPix: Multi-Modal Large Language Model for Pixel-level image Understanding in remote sensing

arXiv

引用

arXiv 2025年

作者： Ou, Ruizhe Hu, Yuan Zhang, Fan Chen, Jiaxin Liu, Yu Pattern Recognition and Intelligent System Lab School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China Institute of Remote Sensing and Geographic Information Systems School of Earth and Space Sciences Peking University Beijing100871 China Peking University Ordos Research Institute of Energy Ordos017000 China

Multi-modal large language models (MLLMs) have achieved remarkable success in image- and region-level remote sensing (RS) image understanding tasks, such as image captioning, visual question answering, and visual grounding. However, existing RS MLLMs lack the pixel-level dialogue capability, which involves responding to user instructions with segmentation masks for specific instances. In this paper, we propose GeoPix, a RS MLLM that extends image understanding capabilities to the pixel level. This is achieved by equipping the MLLM with a mask predictor, which transforms visual features from the vision encoder into masks conditioned on the LLM’s segmentation token embeddings. To facilitate the segmentation of multi-scale objects in RS imagery, a class-wise learnable memory module is integrated into the mask predictor to capture and store class-wise geo-context at the instance level across the entire dataset. In addition, to address the absence of large-scale datasets for training pixel-level RS MLLMs, we construct the GeoPixInstruct dataset, comprising 65,463 images and 140,412 instances, with each instance annotated with text descriptions, bounding boxes, and masks. Furthermore, we develop a two-stage training strategy to balance the distinct requirements of text generation and masks prediction in multi-modal multi-task optimization. Extensive experiments verify the effectiveness and superiority of GeoPix in pixel-level segmentation tasks, while also maintaining competitive performance in image- and region-level benchmarks. The models, dataset, and code are publicly available at https://***/Norman-Ou/GeoPix. © 2025, CC BY.

关键词： image segmentation

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：