检索结果-内蒙古大学图书馆

IEEE International Symposium on Geoscience and remote sensing (IGARSS)

作者： Nora Gourmelon Julian Klink Thorsten Seehaus Matthias Braun Andreas Maier Vincent Christlein Computer Science Department Pattern Recognition Lab Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany Department of Geography and Geosciences Institut für Geographie Friedrich-Alexander-Universität Erlangen-Nürnberg Erlangen Germany

Big advancements in the field of Deep Learning allow the automated extraction of glacier calving fronts from satellite imagery. However, current efforts on Synthetic Aperture Radar (SAR) imagery still produce coarse and partly spurious predictions. Therefore, in this study, a two-dimensional fully-connected Conditional Random Field (CRF) is incorporated into the post-processing of a deep learning-based calving front delineation pipeline. The CRF takes as input the artificial neural network prediction and the original SAR image. Experiments are undertaken using the newly introduced benchmark dataset CaFFe and results are compared to the associated baseline. By introducing the CRF into the post-processing of the baseline’s pipeline, the mean distance error of the front prediction is improved by an average of 27 meters. The code is available at https://***/EntChanelt/GlacierCRF.

关键词：

来源：评论

学校读者我要写书评

暂无评论

11th Pacific-Rim Symposium on image and Video Technology, PSIVT 2023

11th Pacific-Rim Symposium on Image and Video Technology, PS...

引用

11th Pacific-Rim Symposium on image and Video Technology, PSIVT 2023

ISBN: (纸本)9789819703753

The proceedings contain 34 papers. The special focus in this conference is on image and Video Technology. The topics include: Spatial Variation Sequences for remote sensing Applications with Small Sample Sizes;exploring the Potential of High-Resolution Drone imagery for Improved 3D Human Avatar Reconstruction: A Comparative Study with Mobile images;point Cloud Novelty Detection Based on Latent Representations of a General Feature Extractor;Efficient 3Dconv Fusion of RGB and Optical Flow for Dynamic Hand Gesture recognition and Localization;an Investigation of Video Vision Transformers for Depression Severity Estimation from Facial Video Data;real-Time Automated Body Condition Scoring of Dairy Cows;Logo-SSL: Self-supervised Learning with Self-attention for Efficient Logo Detection;HAHANet: Towards Accurate image Classifiers with Less Parameters;evaluating Mammogram image Classification: Impact of Model Architectures, Pretraining, and Finetuning;melanoma Classification Using Deep Learning;3D Formation Control of Multiple Cooperating Autonomous Agents via Leader-Follower Strategy;LAPRNet: Lightweight Airborne Particle Removal Network for LiDAR Point Clouds;REAL-NET: A Monochromatic Depth Estimation Using REgional Attention and Local Feature Mapping;Spike-EFI: Spiking Neural Network for Event-Based Video Frame Interpolation;scrambleMix: A Privacy-Preserving image processing for Edge-Cloud Machine Learning;Comparison of Simplified SE-ResNet and SE-DenseNet for Micro-Expression Classification;facial Deepfake Detection Using Gaussian Processes;A Novel Steganography Scheme Using Logistic Map, BRISK Descriptor, and K-Means Clustering;a Holistic Approach to Elderly Safety: Sensor Fusion, Fall Detection, and Privacy-Preserving Techniques;cluster-Based Video Summarization with Temporal Context Awareness;On Deploying Mobile Deep Learning to Segment COVID-19 PCR Test Tube images;enhancing Safety During Surgical Procedures with Computer Vision, Artificial Intelligence, and Natural

关键词：

来源：评论

学校读者我要写书评

暂无评论

FPGA design of video processing circuit based on 4Kx4K Large array infrared detector 7

FPGA design of video processing circuit based on 4Kx4K Large...

引用

7th International Conference on Electromechanical Control Technology and Transportation, ICECTT 2022

作者： Wang, Lei Huang, Yi Shen, YiWei Yu, Dong Geng, Shufang Liu, Ping Beijing Institute of Space Mechanics & Electricity Beijing China

ISBN: (纸本)9781510656659

As the core component of the camera, infrared focal plane detector assembly is widely used in the field of aerospace infrared remote sensing because of its excellent performance in thermal image viewing, forward-looking early warning, prevention and control monitoring, infrared recognition and so on. Aiming at the 4kx4k large area array infrared video circuit spliced by four 2kx2k infrared detectors, firstly this paper considers the difficulties of piecewise adjustment of integration time and sampling point position delay compensation of spliced detectors from the perspective of FPGA. Then the design idea is given, including configuring and using management chip to reduce clock jitter and improve the reliability of data transmission;the functions of programmable adjustment of pixel blind coordinates and real-time inconsistency correction are realized, and the image quality is improved;using DDR2 to store inconsistency correction parameters;finally, the channel image data synthesis and transmission are completed. All functions such as focal plane control, signal processing and plug-in chip control are completed with one FPGA, which has the advantages of miniaturization and lightweight. The test shows that the design idea and scheme in this paper are feasible and practical, which provides reliable technical support for video signal processing of aerospace remote sensing large array detector, and is of great significance. © 2022 SPIE

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Convolutional Radon Transformation Method for Sparse Collective Recognizing 9

Convolutional Radon Transformation Method for Sparse Collect...

引用

9th International Conference on Control, Automation and Robotics (ICCAR)

作者： Cheng, Bowen Wang, Dan Niu, Jiaxiang Li, Xiaoda Luan, Shenshen Beijing Inst Spacecraft Syst Engn Beijing Peoples R China

ISBN: (纸本)9798350322514

The onboard processing technology based on remote sensing images has become the focus of current research area and received widespread attention. With the increasing amount of onboard data, how to extract effective information from massive data becomes increasingly important. In this paper, a novel sparse collective formation recognizing method is proposed to detect and analyze the formation of the surface ship formation. Firstly, a mobilenet is trained to detect the carrier and single ships to form a binary encoded map, which will be segmented by the DBSCAN method to extract the suspected formation. Then the binary encoded map is processed by the convolutional radon transformation and the result is compared with the standard confidential dataset (SCD) of three formations. Finally the formation type, confidential probability, direction of the surface ship formation was given by the comparison result which has the minimum root-mean-square-error (RMSE) with the SCD. The experimental results show that the proposed method can recognize the 'Y', 'T' and 'I' type surface ship formations with different offsets under more than 60% confidential probability and has a significant feasibility and robustness.

关键词： surface ship formation convolutional radon transformation pattern recognition semantic analyze

来源：评论

学校读者我要写书评

暂无评论

Computer Vision-Based Bengali Sign Language To Text Generation 5

Computer Vision-Based Bengali Sign Language To Text Generati...

引用

5th IEEE International Conference on image processing Applications and Systems (IPAS)

作者： Tazalli, Tonjih Aunshu, Zarin Anan Liya, Sumaya Sadbeen Hossain, Magfirah Mehjabeen, Zareen Ahmed, Md. Sabbir Hossain, Muhammad Iqbal BRAC Univ Dept CSE Dhaka Bangladesh

ISBN: (纸本)9781665462198

In the whole world, around 7% of people have hearing and speech impairment problems. They use sign language as their communication method. As for our country, there are lots of people born with hearing and speech impairment problems. Therefore, our primary focus is to work for those people by converting Bangla sign language into text. There are already various projects on Bangla sign language done by other people. However, they focused more on the separate alphabets and numerical numbers. That is why, we want to concentrate on Bangla word signs since communication is done using words or phrases rather than alphabets. There is no proper database for Bangla word sign language, so we want to make a database for our work using BDSL. In recognition of sign language (SLR), there usually are two types of scenarios: isolated SLR, which takes words by word and completes recognize action, and the other one is continuous SLR, which completes action by translating the whole sentence at once. We are working on isolated SLR. We introduce a method where we are going to use PyTorch and YOLOv5 for a video classification model to convert Bangla sign language into the text from the video where each video has only one sign language word. Here,we have achieved an accuracy rate of 76.29% on the training dataset and 51.44% on the testing dataset. We are working to build a system that will make it easier for hearing and speech-disabled people to interact with the general public.

关键词： Bangla Sign Language Convolutional Neural Network (CNN) Video Classification PyTorch YOLOv5 image processing

来源：评论

学校读者我要写书评

暂无评论

Towards fast sensing along ultralong BOTDA: flatness enhancement by utilizing injection-locked dual-bandwidth probe wave

引用

OPTICS EXPRESS 2022年第12期30卷 20501-20514页

作者： Yang, Yulian Liu, Liming Deng, Qingxue Jia, Xinhong Wu, Han Liang, Wenyan Jiang, Li Song, Weijie Ma, Huiliang Lin, Jiabing Xu, Shirong Sichuan Normal Univ Coll Phys & Elect Engn Chengdu 610101 Sichuan Peoples R China Sichuan Univ Coll Elect Informat Chengdu 610064 Sichuan Peoples R China

Brillouin optical time-domain analysis (BOTDA) using distributed Brillouin amplification (DBA) only requires a milliwatt-level pump to achieve a sensing range beyond 100 km, which provides a powerful tool for temperature/strain sensing. However, similar to the majority of other long-range BOTDAs, the state-of-the-art reports require > 1000 times average, severely restricting the sensing speed. The blind area over tens of kilometers caused by the nonuniform Brillouin response and parasitic amplitude modulation (AM) are crucial factors affecting the signal-to-noise ratio (SNR). Here, a comprehensive performance optimization and substantial enhancement for BOTDA sensors was presented by the direct demodulation of an injection-locked dual-bandwidth probe wave. Injection locking (IL) can completely eliminate the impact of AM noise;dual-bandwidth probe enables self-adaptive pulse loss compensation, thereby intensifying the SNR flatness along the ultralong fiber, and direct probe demodulation can overcome nonlocal effects and allows similar to 19.7 dB enhancement of probe input power. Therefore, using only 100 times average, similar to 148.3 km sensing, and similar to 5 m spatial resolution were achieved with < similar to 0.8 MHz standard deviation of Brillouin frequency shift (BFS) over a broad range (similar to 131.71cm). The reduction in averages was more than 10 times that of the reported majority of long-range BOTDAs. Such performances were achieved without using time-consuming or post-processing techniques, such as optical pulse coding and image denoising. Because this approach is compatible with optical chirp chain technique without frequency sweeping, fast acquisition (0.3 s) was also realized, which has the potential for fast sensing at 3.3 Hz along a similar to 150 km fiber. (C) 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

关键词： Optical signal to noise ratio pattern recognition Self phase modulation Spatial resolution Stimulated Raman scattering Stimulated scattering

来源：评论

学校读者我要写书评

暂无评论

Adaptive Fine-Tuning and Feature Enhancement of SAM for remote sensing Instance Segmentation

Adaptive Fine-Tuning and Feature Enhancement of SAM for Remo...

引用

Symposium of image, Signal processing, and Artificial Vision (STSIVA)

作者： Zhixin Zhao Yangguang Liu Lingwu Meng Haochen Zhang Mingyang Zhu Liang Xiao School of Computer Science and Technology Nanjing University of Science and Technology Nanjing China

ISBN: (数字)9798350363203

ISBN: (纸本)9798350363210

remote sensing image instance segmentation aims to accurately separate independent objects and automatically identify land attributes. Recent advancements in large models have propelled self-supervised learning, especially with the segment anything model (SAM). Although SAM can generate instance segmentation masks without additional training, its application to remote sensing images still faces challenges, such as a severe reliance on manual prior prompts, inaccuracies in detail segmentation, and insufficient generalization capability. Therefore, to enhance the generalization capability of SAM in the field of remote sensing images, we propose a SAM-based fine-tuning model called AFFE-SAM. This model combines adaptive low-rank (AdaLoRA) and variable adapter (MiMi-Adapter) fine-tuning methods. Additionally, it improves mask quality in the decoder by employing adaptive multi-scale feature enhancement. We conducted experiments on two datasets, obtaining mAP scores of 66.5 for the NWPU VHR-10 dataset and 67.1 for the SSDD dataset. The evaluation results validate the effectiveness of our method.

关键词： Instance segmentation Training Adaptation models Self-supervised learning Propulsion Decoding Sensors Object recognition Informatics remote sensing

来源：评论

学校读者我要写书评

暂无评论

Time Dependent image Generation of Plants from Incomplete Sequences with CNN-Transformer 1

引用

44th DAGM German Conference on pattern recognition (DAGM GCPR)

作者： Drees, Lukas Weber, Immanuel Russwurm, Marc Roscher, Ribana Tech Univ Munich Data Sci Earth Observat Munich Germany Univ Bonn Remote Sensing IGG Bonn Germany AB EX PLEdoc Essen Germany Ecole Polytech Fed Lausanne EPFL ECEO Lab Lausanne Switzerland

ISBN: (数字)9783031167881

ISBN: (纸本)9783031167881;9783031167874

Data imputation of incomplete image sequences is an essential prerequisite for analyzing and monitoring all development stages of plants in precision agriculture. For this purpose, we propose a conditional Wasserstein generative adversarial network TransGrow that combines convolutions for spatial modeling and a transformer for temporal modeling, enabling time-dependent image generation of above-ground plant phenotypes. Thereby, we achieve the following advantages over comparable data imputation approaches: (1) The model is conditioned by an incomplete image sequence of arbitrary length, the input time points, and the requested output time point, allowing multiple growth stages to be generated in a targeted manner;(2) By considering a stochastic component and generating a distribution for each point in time, the uncertainty in plant growth is considered and can be visualized;(3) Besides interpolation, also test-extrapolation can be performed to generate future plant growth stages. Experiments based on two datasets of different complexity levels are presented: Laboratory single plant sequences with Arabidopsis thaliana and agricultural drone image sequences showing crop mixtures. When comparing TransGrow to interpolation in image space, variational, and adversarial autoencoder, it demonstrates significant improvements in image quality, measured by multi-scale structural similarity, peak signal-to-noise ratio, and Frechet inception distance. To our knowledge, TransGrow is the first approach for time- and image-dependent, high-quality generation of plant images based on incomplete sequences.

关键词： Data imputation Transformer Positional encoding image time series Conditional GAN image generation Plant growth modeling

来源：评论

学校读者我要写书评

暂无评论

Small Object Detection Leveraging on Simultaneous Super-resolution 25

Small Object Detection Leveraging on Simultaneous Super-reso...

引用

25th International Conference on pattern recognition (ICPR)

作者： Ji, Hong Gao, Zhi Liu, Xiaodong Zhang, Yongjun Mei, Tiancan Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430079 Peoples R China Natl Univ Singapore Dept Elect & Comp Engn Singapore 117583 Singapore Wuhan Univ Elect & Informat Sch Wuhan 430072 Peoples R China

ISBN: (纸本)9781728188089

Despite the impressive advancement achieved in object detection, the detection performance of small object is still far from satisfactory due to the lack of sufficient detailed appearance to distinguish it from similar objects. Inspired by the positive effects of super-resolution for object detection, we propose a framework that can be incorporated with detector networks to improve the performance of small object detection, in which the low-resolution image is super-resolved via generative adversarial network (GAN) in an unsupervised manner. In our method, the super-resolution network and the detection network are trained jointly. In particular, the detection loss is back-propagated into the super-resolution network during training to facilitate detection. Compared with available simultaneous super-resolution and detection methods which heavily rely on low-/high-resolution image pairs, our work breaks through such restriction via applying the CycleGAN strategy, achieving increased generality and applicability, while remaining an elegant structure. Extensive experiments on datasets from both computer vision and remote sensing communities demonstrate that our method obtains competitive performance on a wide range of complex scenarios.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Target Classification and recognition for High-Resolution remote sensing images: Using the Parallel Cross-Modal Neural Cognitive Computing Algorithm

引用

IEEE GEOSCIENCE AND remote sensing MAGAZINE 2020年第3期8卷 50-62页

作者： Liu, Yang Xie, Yi Yang, Wei Zuo, Xianyu Ge, Qiang Zhou, Bing Henan Univ Coll Comp Sci & Informat Engn Kaifeng Peoples R China Henan Univ Coll Environm & Planning Kaifeng Peoples R China Henan Univ Kaifeng Peoples R China

Target classification and recognition (TCR) are important information-extraction techniques for high-resolution remote sensing images (HRIs). However, because methods with high accuracy usually have higher time complexity, the massive remote sensing image has brought great difficulties for realtime application. In this article, we propose a hybrid, heterogeneous parallel processing algorithm to improve the efficiency TCR based on the Crossmodel Neural Cognitive Computing (CNCC) algorithm [parallel TCR based on MapReduce of hierarchical CNCC (PTCR-C)]. Parallel programming technologies used in this article include Open Multiprocessing (OpenMP), Compute Unified Device Architecture (CUDA), and the message passing interface (MPI). Experiments on public remote sensing image data sets show that the PTCR-C algorithm can effectively improve the efficiency of the original TCR algorithm. The PTCR-C algorithm has better augmented ability and provides the reference for further real-time TCR of remote sensing applications.

关键词： remote sensing Parallel processing Graphics processing units Classification algorithms Computational modeling Training Target recognition

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：