检索结果-内蒙古大学图书馆

Factor annealing decoupling compositional training method for imbalanced hyperspectral image classification

IET image processing 2024年第10期18卷 2553-2567页

作者： Li, Xiaojun Su, Yi Yao, Junping Guo, Yi Fan, Shuai Xian Res Inst High Technol Informat Syst Dept Xian Shannxi Peoples R China Xian Res Inst High Technol Xian 710000 Shannxi Peoples R China

Due to differences in the quantity and size of observed targets, hyperspectral images are characterized by class imbalance. The standard deep learning classification model training scheme optimizes the overall classification error, which may lead to performance imbalance between classes in hyperspectral image classification frameworks. Therefore, a novel factor annealing decoupling compositional training method is proposed in this paper. Without requiring resampling or reweighting, it implicitly modulates the training process, so standard models can sufficiently learn the representation of the minority classes and further be trained as robust classifiers. Specifically, the label-distribution-aware margin loss is combined with the error-rate-based cross-entropy loss via combination factor, which considers both imbalanced data representation learning and classifier overall performance. Then, a factor annealing optimization training scheme is designed to adjust the combination factor, which solves the stage division problem of two-stage decoupling learning. Experimental results on two hyperspectral image datasets demonstrate that, as compared with other competing approaches, the proposed method can continuously and stably optimize the model parameters, achieving improvements in class average metrics and difficult classes without affecting overall classification performance. A novel factor annealing decoupling compositional training method for imbalanced hyperspectral image classification is proposed in this paper. It considers both imbalanced data representation learning and classifier overall performance and solves the stage division problem of two-stage decoupling learning. image

关键词： image classification image processing image representation learning (artificial intelligence) pattern classification remote sensing

来源：评论

学校读者我要写书评

暂无评论

Fine-Grained Species recognition With Privileged Pooling: Better Sample Efficiency Through Supervised Attention

引用

IEEE TRANSACTIONS ON pattern ANALYSIS AND MACHINE INTELLIGENCE 2023年第12期45卷 14575-14589页

作者： Rodriguez, Andres C. D'Aronco, Stefano Schindler, Konrad Wegner, Jan Dirk Swiss Fed Inst Technol EcoVis Lab Photogrammetry & Remote Sensing CH-8092 Zurich Switzerland Univ Zurich Inst Computat Sci CH-8006 Zurich Switzerland

We propose a scheme for supervised image classification that uses privileged information, in the form of keypoint annotations for the training data, to learn strong models from small and/or biased training sets. Our main motivation is the recognition of animal species for ecological applications such as biodiversity modelling, which is challenging because of long-tailed species distributions due to rare species, and strong dataset biases such as repetitive scene background in camera traps. To counteract these challenges, we propose a visual attention mechanism that is supervised via keypoint annotations that highlight important object parts. This privileged information, implemented as a novel privileged pooling operation, is only required during training and helps the model to focus on regions that are discriminative. In experiments with three different animal species datasets, we show that deep networks with privileged pooling can use small training sets more efficiently and generalize better.

关键词： Camera trap images fine-grained species recognition privileged pooling supervised attention training set bias

来源：评论

学校读者我要写书评

暂无评论

Application of remote sensing image processing for Classification and recognition

Application of Remote Sensing Image Processing for Classific...

引用

International Conference on Advanced Infocomm Technology (ICAIT)

作者： Xiaolong Shi Dan Huang Wenjie Li Xianjie Wang College of Computer Science and Engineering Chongqing University of Technology Chongqing China China Research and Development Academy of Machinery Equipment Beijing China Chongqing high-tech Zone Pegasus Innovation Institute Chongqing University of Technology Chongqing China

Aiming at the difficulties in object detection and recognition in remote sensing images caused by high background complexity, large scale variations of targets, and the presence of numerous small objects, an improved method for remote sensing image object detection based on YOLOv7-tiny is proposed. This method combines the loss function based on normalized Gaussian Wasserstein distance (NWD) with the CIoU loss function to address the problem of sensitivity to positional deviation of small objects by IoU-Loss. The addition of a global attention mechanism (GAM) in the backbone network reduces information diffusion and enhances the interaction at the global dimension to mitigate the interference of complex backgrounds in remote sensing images on the model, enabling the model to focus on the feature extraction of the desired targets. Finally, the coupled detection head (Coupled Head) of the model is replaced with a decoupled detection head (Decoupled Head), allowing the classification and regression tasks to output from different branches to achieve decoupling and avoid the decrease in detection accuracy caused by conflicts between classification and regression. The experimental results of this method on the public dataset DIOR achieved 88.73% accuracy, which is an improvement of 1.78% compared to the unimproved method's accuracy of 86.95%. Furthermore, compared to other researchers' methods tested on DIOR, the proposed method also shows improvement, thus validating its effectiveness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CSVD: a cross-scenario vehicle dataset for multi-object tracking

CSVD: a cross-scenario vehicle dataset for multi-object trac...

引用

2024 International Conference on image, Signal processing, and pattern recognition, ISPP 2024

作者： Li, Xiaolei Zhou, Juefan Xiao, Xingjie Lin, Jiayu Yang, Siyuan Sha, Zongyao Tu, Jianguang School of Remote Sensing and Information Engineering Wuhan University Hubei Province 430079 China Wuhan ALLPRS Remote Sensing Data Technology Co. Ltd. Hubei Province 430079 China

ISBN: (纸本)9781510680425

Leveraging visual sensing technologies for the detection and tracking of vehicles represents a critical application domain for unmanned aerial vehicles (UAVs), notably in challenging operational *** study focuses on enhancing UAV functionalities in intricate environments through the development of a specialized dataset, derived from battlefield scenarios, to facilitate advanced research on vehicle detection and multi-target tracking under complex conditions.A comprehensive collection of vehicular movement videos spanning diverse scenarios was amassed and manually annotated, culminating in the creation of the " Cross-Scenario Vehicle Detection" (CSVD) *** dataset encompasses a wide array of environmental settings, featuring urban landscapes, plains, and forests, across the four seasons, resulting in a total of 13,025 meticulously annotated *** several state-of-the-art deep learning models, we established robust benchmarks for object ***, an extensive evaluation and performance validation were conducted using cutting-edge multi-object tracking algorithms on the CSVD dataset, incorporating diverse assessment *** conducted experiments demonstrate the dataset's robust applicability and versatility, endorsing its effectiveness for the development and evaluation of UAV-based vehicle detection and multi-target tracking systems in complex settings. © 2024 SPIE.

关键词： Aircraft detection

来源：评论

学校读者我要写书评

暂无评论

Attribute-Guided Generative Adversarial Network With Improved Episode Training Strategy for Few-Shot SAR image Generation

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2023年 16卷 1785-1801页

作者： Sun, Yuanshuang Wang, Yinghua Hu, Liping Huang, Yuanyuan Liu, Hongwei Wang, Siyuan Zhang, Chen Xidian Univ Natl Lab Radar Signal Proc Xian 710071 Peoples R China Beijing Inst Environm Features Sci & Technol Electromagnet Scattering Lab Beijing 100854 Peoples R China

Deep-learning-based models usually require a large amount of data for training, which guarantees the effectiveness of the trained model. Generative models are no exception, and sufficient training data are necessary for the diversity of generated images. However, for synthetic aperture radar (SAR) images, data acquisition is expensive. Therefore, SAR image generation under a few training samples is still a challenging problem to be solved. In this article, we propose an attribute-guided generative adversarial network (AGGAN) with an improved episode training strategy for few-shot SAR image generation. First, we design the AGGAN structure, and spectral normalization is used to stabilize the training in the few-shot situation. The attribute labels of AGGAN are designed to be the category and aspect angle labels, which are essential information for SAR images. Second, an improved episode training strategy is proposed according to the characteristics of the few-shot generative task, and it can improve the quality of generated images in the few-shot situation. In addition, we explore the effectiveness of the proposed method when using different auxiliary data for training and use the Moving and Stationary Target Acquisition and recognition benchmark dataset and a simulated SAR dataset for verification. The experimental results show that AGGAN and the proposed improved episode training strategy can generate images of better quality when compared with some existing methods, which have been verified through visual observation, image similarity measures, and recognition experiments. When applying the generated images to the 5-shot SAR image recognition problem, the average recognition accuracy can be improved by at least 4$\%$.

关键词： Few-shot image generation generative adversarial network (GAN) meta-learning synthetic aperture radar (SAR) transfer learning

来源：评论

学校读者我要写书评

暂无评论

Spatial planning method of urban landscape architecture distribution pattern based on evolutionary algorithm

引用

INTERNATIONAL JOURNAL OF ENVIRONMENTAL TECHNOLOGY AND MANAGEMENT 2024年第1-2期27卷 66-78页

作者： Feng, Jingjing Chen, Yuxiong Guangdong Nanfang Inst Technol Sch Informat Jiangmen 529000 Peoples R China

Aiming at the problems of low planning accuracy and long planning time in the traditional spatial planning method of urban landscape architecture distribution pattern, a spatial planning method of urban landscape architecture distribution pattern based on evolutionary algorithm was proposed. First, we acquire urban landscape remote sensing images through ETM+ and Landsat TM/OLI images, and use ENVI software to conduct geometric correction, image enhancement and other image processing. Then, we acquire spatial data of landscape distribution pattern from urban landscape green space types, patch area size, number and other aspects. We then use differential evolution algorithm to calculate the fitness value corresponding to the initialised population, extract landscape features, and use mutation operators. The optimal solution is obtained through the three steps of crossover operator and selection operation, which is the optimal spatial planning strategy. The simulation results show that the proposed method has higher precision and shorter planning time in spatial planning of urban landscape architecture distribution pattern.

关键词： evolutionary algorithm urban landscape architecture distribution pattern spatial planning landscape characteristics

来源：评论

学校读者我要写书评

暂无评论

A Novel Real-Time Text-to-Speech System Using Raspberry Pi for Assisting the Visually Impaired

引用

TRAITEMENT DU SIGNAL 2024年第6期41卷 3183-3192页

作者： Ben Atitallah, Ahmed Kammoun, Manel Atitallah, Mohamed Amin Ben Albekairi, Mohammed Said, Yahia Boudabous, Anis Kaaniche, Khaled Atri, Mohamed Jouf Univ Coll Engn Dept Elect Engn Sakaka 72388 Saudi Arabia Univ Sfax LETI ENIS Sfax 3029 Tunisia Gustave Eiffel Univ Lab Informat Gaspard Monge CNRS A3SIESIEE Paris BP 99 F-93162 Noisy Le Grand France Northern Border Univ Coll Engn Remote Sensing Unit Ar Ar 91431 Saudi Arabia Univ Monastir Lab Elect & Microelect LR99ES30 Monastir 5019 Tunisia Jouf Univ Coll Comp & Informat Sci Dept Comp Engn & Networks Sakaka 72388 Saudi Arabia King Khalid Univ Coll Comp Sci Abha 62529 Saudi Arabia

Visual impairment is one of the most significant challenges facing humanity, Aespecially in an era where information is frequently conveyed through text rather than voice. To address this, the proposed system is designed to assist individuals with visual impairments. This paper presents the development of a real-time Text-to-Speech (TTS) Aembedded system based on the Raspberry Pi 4. AOur system incorporates a novel approach to enhance the accuracy of text recognition using Optical Character recognition (OCR) from images. Specifically, a series of preprocessing steps are employed, selected dynamically by a decision-making process based on the content of the image. The image processing is handled using OpenCV2, while the conversion of text to speech is achieved through the pyttsx3 Python library. The entire system is implemented and tested on a Raspberry Pi 4, connected to a USB Full HD camera for high-resolution image acquisition, and controlled via the Traffic HAT-LED module. Experimental results demonstrate that our system achieves a minimum accuracy of 88.33% in text recognition from images.

关键词： image preprocessing visual impairment Raspberry Pi 4 text-to-speech optical character recognition real-time processing

来源：评论

学校读者我要写书评

暂无评论

Research on target recognition based on optical neural networks

Research on target recognition based on optical neural netwo...

引用

2024 International Conference on remote sensing, Mapping, and image processing, RSMIP 2024

作者： Zhang, Yixuan Feng, Yuxiang Chang, Hong Beijing Aerospace Institute for Metrology and Measurement Technology Beijing China

ISBN: (纸本)9781510680012

Due to the advantages of high throughput, low latency, and low power consumption, optical neural networks hold great promise in addressing the challenges of energy consumption and computational efficiency faced by current artificial intelligence technologies. Consequently, they have become a research hotspot in both academia and industry in recent years. The goal of optical neural networks is to use photons as the physical carrier to construct the basic computational units of artificial neural network algorithms, thus achieving high-performance novel computing architectures and applying them to solve practical problems. This paper introduces the working principles and characteristics of optical neural networks and discusses relevant research on target recognition based on optical neural network architectures. © 2024 SPIE.

关键词： Optical data processing

来源：评论

学校读者我要写书评

暂无评论

Research on Object-Oriented Classification Technology for remote sensing imagery of Coastal Zone 11th

Research on Object-Oriented Classification Technology for Re...

引用

11th International Conference on Signal and Information processing, Network and Computers, ICSINC 2023

作者： Yize, Dong Rui, Zhang Haitao, Wang Chao, Wang Xianglei, Kong Lele, Yao Institute of Spacecraft Application System Engineering CAST No. 104 Youyi Road Haidian Beijing China

ISBN: (纸本)9789819721191

Terrain identification of coastal is of great significance for coastal development activities and coastal terrain survey in overseas areas. However, due to the complex characteristics of coastal features, the use of remote sensing images for automatic feature classification and recognition has become a current research hotspot. Utilizing the homogeneous and homogeneous spectral characteristics of hyperspectral remote sensing images and the characteristics of coastal zone elements, an object-oriented shoreland classification technique is proposed after performing operations such as atmospheric correction and image enhancement preprocessing on hyperspectral remote sensing data, which enables automatic identification and extraction of feature types and geomorphological information in the coastal zone. This method overcomes the limitation of the traditional hyperspectral remote sensing classification method, which takes image element as the processing unit, and synthesizes the spatial characteristics and spectral characteristics of the features, which greatly reduces the classification spots and significantly improves the classification effect. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Coastal zones

来源：评论

学校读者我要写书评

暂无评论

The Research of Facial Expression image recognition Method Based on MobileNetV3

The Research of Facial Expression Image Recognition Method B...

引用

2024 International Conference on remote sensing, Mapping, and image processing, RSMIP 2024

作者： Zou, Xinyue Liu, Chenguang Xu, Xuebin Zhang, Rong School of Computer Science and Technology Xi'an University of Posts &Telecommunications Shaanxi Xi'an China China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Shaanxi Xi'an China

ISBN: (纸本)9781510680012

The information conveyed through facial expressions accounts for a large proportion of the total information and can effectively express people's intentions and emotions. Facial expression recognition has laid the foundation for fields such as human-computer interaction, facial emotion prediction, and artificial intelligence, and has become an important research object in computer vision. This article proposes a facial expression recognition method based on the MobileNetV3 network for face images from different angles. The method uses depth-wise separable convolution, introduces attention mechanism and new activation function to update blocks, and redesigns the time-consuming layer structure at the end. The dataset used in this article is the KDEF, which includes 4, 900 color images with a size of 562*762 pixels. Through extensive experiments, it has been shown that the proposed structure in this article improves the accuracy of facial expression recognition from different angles compared to other network structures, achieving 94.7%, and has a smaller parameter count, which is beneficial for further research on facial expressions. © 2024 SPIE.

关键词： image classification

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：