检索结果-内蒙古大学图书馆

An Effective Feature Extraction Approach Based on Spectral-Gabor Space Discriminant Analysis for Hyperspectral image

引用

NEURAL processing LETTERS 2022年第2期54卷 909-959页

作者： Li, Li Gao, Jianqiang Ge, Hongwei Zhang, Yixin Yang, Jieming Jining Med Univ Sch Med Informat Engn Rizhao 276826 Shandong Peoples R China Jiangnan Univ Jiangsu Prov Engn Lab Pattern Recognit & Computat Wuxi 214122 Jiangsu Peoples R China

With the development of remote sensing technology, hyperspectral sensors can record reflectance of ground objects in hundreds of bands, which undoubtedly brings great benefits for hyperspectral remote sensing image that contains abundant information of the land covers. At present, some hyperspectral images are contaminated by the noise and a lack of spatial information, this phenomenon hinders the improvement of classification accuracy. To further mine the spatial features of hyperspectral image, we propose an effective feature extraction method based on Spectral-Gabor space discriminant analysis (SGDA). The framework of SGDA is roughly divided into four steps: firstly, we obtain p(i){p(i)vertical bar 1 <= i < d} principal components by PCA, where d denotes the number of features;secondly, we filter the p(i)principal components with Gabor filter on five different scales and eight different orientations and obtain Gabor spatial features;thirdly, incorporating the spectral features of original hyperspectral data into the Gabor spatial features to form the Spectral-Gabor space features F;finally, we project the fusion features to a low-dimensional subspace, and then maximize the Spectral-Gabor space between-class scatter matrix (S-b(SG)) and minimize the Spectral-Gabor space within-class scatter matrix (S-w(SG)) at the same time inspired by the idea of Fisher line discriminant analysis. Also, in the above fusion process, the proportion of spectral and spatial information can be controlled by using a penalty factor alpha. Experimental results on three hyperspectral datasets show the better classification performance of the SGDA method than the state-of-art methods on the condition of small sample size by Maximum Likelihood Classifier (MLC).

关键词： Hyperspectral image Feature extraction Spectral-Gabor space discriminant analysis Classification

来源：评论

学校读者我要写书评

暂无评论

ICVIP 2023 - Proceedings of 2023 7th International Conference on Video and image processing

ICVIP 2023 - Proceedings of 2023 7th International Conferenc...

引用

7th International Conference on Video and image processing, ICVIP 2023

ISBN: (纸本)9798400709388

The proceedings contain 12 papers. The topics discussed include: image classification trusted model based on horizontal federated learning and SGX;NAC-TCN: temporal convolutional networks with causal dilated neighborhood attention for emotion understanding;a novel two-stage data-mining model combining gait recognition and temporal sequence mining;underwater image enhancement with color correction using convolutional neural networks;infrared image enhancement algorithm based on multiscale guided filtering;underwater image dehazing in YCbCr color space using superpixel segmentation;a vehicle detection method under strong infrared radiation interference: using YOLOv8 to recognize cars in infrared images;super-resolution reconstruction of remote sensing image by fusion of receptive field and attention;and safety helmet wearing detection algorithm based on lightweight FastestDet.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hardware-Level image recognition System Based on ZnO Photo-Synapse Array with the Self-Denoising Function

引用

ADVANCED FUNCTIONAL MATERIALS 2024年第19期34卷 2313507-2313507页

作者： Jiang, Jiandong Xiao, Wei Li, Xiang Zhao, Yanfei Qin, Zhaoyang Xie, Zhichao Shen, Guangyue Zhou, Jianhong Fu, Yujun Wang, Yanrong Wang, Qi He, Deyan Lanzhou Univ Sch Mat & Energy Lanzhou 730000 Peoples R China

The emerging optoelectronic neuromorphic devices are widely concerned due to their capability to integrate the functions of signal sensing, memory, and processing. Although significant advancements have been made in the study of individual optoelectronic synaptic devices, the development of hardware-level image recognition systems based on photo-synapse arrays remains a challenge. In this study, a crosstalk-free, easy-to-integrate, and scalable 8 x 8 crossbar array for optical image sensing and storage is demonstrated using vertical two-terminal ZnO photo-synapses with the self-denoising function. By designing peripheral circuits, a complete hardware-level artificial visual system is constructed that successfully implements the real-time pattern recognition tasks for 8 x 8 pixel images. The excellent performance of the photo-synapse array shows its remarkable ability in highly efficient optic neuromorphic computing. Additionally, an in-sensor reservoir computing (RC) system is constructed for image recognition of handwritten digits. The system achieves a high classification accuracy of 95.1%. In this study, an 8x8 crossed array is demonstrated. The array consists of 64 vertical two-terminal ZnO photo-synapses. The array enables the sensing and storage of optical images. In addition, a complete hardware-level artificial visual system is constructed and successfully implemented a real-time pattern recognition task for 8 x 8 pixel ***

关键词： bionic vision system crossbar array image self-denoising photo-synapse reservoir computing

来源：评论

学校读者我要写书评

暂无评论

MSDGAN: Multi-Scale Dilated Generative Adversarial Network for Smoke Removal and image Restoration 24

MSDGAN: Multi-Scale Dilated Generative Adversarial Network f...

引用

2024 5th International Artificial Intelligence and Blockchain Conference, AIBC 2024

作者： Gwak, KyungMIn Rho, Young. J Tech University of Korea Siheung Korea Republic of

ISBN: (纸本)9798400710780

This paper presents a novel approach for smoke removal and image restoration using a Multi-Scale Dilated Generative Adversarial Network (MSDGAN). The presence of smoke in images poses significant challenges to both human perception and computer vision tasks, reducing visibility and complicating image processing tasks such as object recognition and segmentation. Traditional techniques like dehazing and defogging often fail to effectively address the irregular and transparent nature of smoke. To overcome these limitations, we propose a network architecture that integrates multi-scale dilated convolutions within a GAN framework, enabling the model to capture contextual information across various scales and improve the restoration of images affected by smoke. The MSDGAN architecture includes an encoder-decoder structure with skip connections, and it leverages perceptual loss and SSIM loss to maintain feature fidelity and reconstruct high-quality images. Extensive experiments demonstrate the effectiveness of the proposed network in removing smoke and restoring image quality across various challenging environments. The results show that MSDGAN outperforms existing dehazing and image restoration methods, offering superior performance as measured by L1, L2, SSIM, PSNR, and LPIPS metrics. This work provides a robust solution for enhancing image clarity in the presence of smoke, with potential applications in fields such as autonomous driving, surveillance, and remote sensing. Copyright © 2024 held by the owner/author(s). Publication rights licensed to ACM.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Research on Target recognition Based on Deep Feature Fusion of Infrared and Visible Light images

Research on Target Recognition Based on Deep Feature Fusion ...

引用

Signal processing, Communications and Engineering Management Conference (ISPCEM), International

作者： Jiahao Wang Juan Ren Jianjie Wang Xiuping Wang Department of Electronic and Information Engineering Zhejiang Gongshang University Hangzhou Zhejiang China Hangzhou Normal University Zhejiang Hangzhou China Xi'an University of Electronic Science and Technology Xi’an China

ISBN: (数字)9798331528676

ISBN: (纸本)9798331528683

It is expected that unmanned aerial vehicle remote sensing target recognition will have a broad range of applications in fields such as smart cities, traffic monitoring, and disaster monitoring. However, remote sensing images have relatively small targets, and they are also affected by factors such as weather, lighting, and occlusion. This makes target detection and recognition difficult. In order to improve the accuracy of target recognition, this paper proposes a dual modal image depth feature fusion method for recognizing unmanned aerial vehicle remote sensor images. The method includes dual modal image registration, and a dual modal image depth feature fusion method based on the yolov7 model. By constructing a bimodal dataset and using the HOPC operator for image registration, strict registration was achieved with manual annotation assistance. The yolov7-dual model proposed in this article, through pixel level fusion and CAFM attention mechanism, ultimately increased the mAP value from 0.424 to 0.657, significantly exceeding expectations.

关键词： image registration Target recognition Smart cities Lighting Object detection Signal processing Autonomous aerial vehicles remote sensing Monitoring Meteorology

来源：评论

学校读者我要写书评

暂无评论

Exploring Advanced image Refinement with Blind Denoising Autoencoder Architecture 2

Exploring Advanced Image Refinement with Blind Denoising Aut...

引用

2nd International Conference on Recent Trends in Microelectronics, Automation, Computing, and Communications Systems, ICMACC 2024

作者： Reddy, V Sagar Reddy, G Venkata Viswas Mohan, K Pradhyuth Raj, A Chetan Sai VNRVJIET Dept of ECE Hyderabad India

ISBN: (纸本)9798350366570

This project introduces an innovative method for enhancing license plate images by employing a blind autoencoder-based denoising and deblurring technique. Unlike conventional approaches that heavily rely on labeled training data, which can be both scarce and costly to obtain, our method circumvents this requirement. Instead, it leverages a blind autoencoder trained on a vast collection of unlabeled license plate images to effectively denoise and deblur input images. Through evaluation on a standard dataset, our approach demonstrates significant enhancements in peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) compared to conventional methods. © 2024 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

RepSViT: An Efficient Vision Transformer Based on Spiking Neural Networks for Object recognition in Satellite On-Orbit remote sensing images

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND remote sensing 2024年 62卷 1页

作者： Pang, Yanhua Yao, Libo Luo, Yiping Dong, Chengguo Kong, Qinglei Chen, Bo Harbin Inst Technol Shenzhen Inst Space Sci & Appl Technol Shenzhen 518055 Peoples R China Naval Aviat Univ Res Inst Informat Fus Yantai 264001 Peoples R China Deep Space Explorat Lab Beijing 100195 Peoples R China Weifang Univ Sci & Technol Sch Architectural Engn Weifang 262700 Peoples R China

The role of on-orbit computing for satellites is transitioning from being a backup measure to becoming a primary key function. However, the limited computing resources available on satellites make it difficult to deploy advanced models with large parameters. Additionally, satellite on-orbit computing requires high speed and accuracy, posing significant challenges for developing suitable models. To overcome these challenges, we propose an efficient vision transformer, RepSViT, for satellite on-orbit computing. The RepSViT introduces spiking neural networks (SNNs) with high biological plausibility, event-driven property, and low power consumption into the field of remote sensing image processing and satellite on-orbit computing for the first time and incorporates structural reparameterization. Specifically, we design a dynamic dilated spiking convolution ((DSC)-S-2) based on SNNs to improve the feature extraction capability and efficiency of RepSViT. We also develop a spiking guided attention module (SGAM) to make RepSViT pay more attention to object-related features with lower computational costs. Furthermore, we design an efficient coupled fine-coarse-grained block (ECFC) to enhance the model's capability in extracting coarse and fine-grained features. To ensure effective feature extraction, inference speed, and reduced computational costs, we design a reparameterized feed-forward network (RepFFN). RepSViT achieves an inference latency of 8.33 ms and a recognition accuracy of 95% on an embedded GPU, utilizing 3.77 million parameters and consuming 0.6 GFLOPs computational costs.

关键词： Feature extraction Satellites Computational modeling Computational efficiency Transformers Neural networks Task analysis Fine-grained object recognition remote sensing image (RSI) satellite-on-orbit computing spiking neural networks (SNNs)

来源：评论

学校读者我要写书评

暂无评论

An outstanding adaptive multi-feature fusion YOLOv3 algorithm for the small target detection in remote sensing images

引用

pattern ANALYSIS AND APPLICATIONS 2022年第4期25卷 951-962页

作者： Li, Guoqiang Hao, Xinyu Zha, Linlin Chen, Anbang Yanshan Univ Sch Elect Engn Qinhuangdao 066004 Hebei Peoples R China Yanshan Univ Key Lab Ind Comp Control Engn Hebei Prov Qinhuangdao 066004 Hebei Peoples R China

Deep learning-based target detection for optical remote sensing images is a significant research direction in the field of image processing. Different from natural images, remote sensing images are characterized by complex backgrounds, similarity of characteristics between various classes and diverse target scales. In this paper, we propose an adaptive multi-feature fusion YOLOv3 remote sensing small target detection algorithm to cope with these features. In the proposed algorithm, the shallow semantic information is extracted by the accessory feature extraction network, and fused with the deep features extracted by Darknet-53 down-sampling to enrich the semantic and spatial information of the feature layers on the auxiliary network. In addition, the shallow features are filtered using the adaptive feature selection module to refine the effective feature information. A cross-layer feature fusion module is proposed to fuse different feature layers to enhance the connection between the semantic information of feature contexts to obtain more information about the characteristics of small targets. To test the effectiveness of the proposed algorithm, it is validated on the Pascal voc2007 dataset. The experimental results show that the detection accuracy of the proposed algorithm could achieve 88.3%, and evidently superior to the original YOLOv3 algorithm. Finally, the proposed algorithm is applied to detect the small target in remote sensing images. The detection results show that compared with the original YOLOv3 algorithm, the mean average precision(mAP) of the proposed algorithm is improved by 2.6%, which can effectively detect more small targets and significantly improve the detection accuracy of small targets than other classical algorithms.

关键词： Small target YOLOv3 Multi-feature fusion Adaptive feature extraction Target detection

来源：评论

学校读者我要写书评

暂无评论

On Real-Time Object recognition by Single image Dehazing Method Using Deep Learning Approach 3

On Real-Time Object Recognition by Single Image Dehazing Met...

引用

3rd International Conference on Advancement in Electronics and Communication Engineering, AECE 2023

作者： Pathak, Shruti Doegar, Amit National Institute of Technical Teachers Training and Research Department of Computer Science Chandigarh160019 India

ISBN: (纸本)9798350330724

In the past decade, various haze removal techniques have been widely reported for object recognition. But hitherto little has been identified on the use of single image dehazing using transfer learning approach for object detection. Single image dehazing is an emerging computer vision technology which offers some of the extreme benefits over the existing techniques such as consumes less processing time, requires less space for real time dehaze purpose, etc. In this study, we combine both the object detection and image dehazing methods for real time applications such as-remote sensing, video surveillances, driverless automatic vehicles, etc. This paper presents an effective and efficient image dehazing method using transfer learning which helps to recognize objects in real time with more clarity and that can automatically detect objects with a high recognition rate and lesser probability of error. Our tests show that object detection becomes less accurate as the haze intensity increases;yet, under all haze circumstances (low, medium, or heavy), our jointly trained model AOD-net +YOLO v3 consistently outperforms non-joint and naïve YOLO v3 techniques. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Approaching semi-supervised collaborative learning model for remote sensing image analysis

Approaching semi-supervised collaborative learning model for...

引用

2022 RIVF International Conference on Computing and Communication Technologies, RIVF 2022

作者： Do, Viet Duc Mai, Dinh Sinh Ngo, Long Thanh Le Quy Don Technical University Dept. Information Technology Hanoi Viet Nam Institute of Techniques for Special Engineering Le Quy Don Technical University Hanoi Viet Nam

ISBN: (纸本)9781665461665

remote sensing image data often has many advantages in mapping, target recognition, object tracking, and so on. However, remote sensing image data often face problems such as big data, multiple dimensions, and time series. These things greatly affect the problem of remote sensing image analysis. In this paper, we extend the collaborative clustering algorithm for remote sensing image analysis by using a fuzzy semi-supervised technique. The proposed algorithm applies in the case of very few labeled data. In particular, the collaborative learning model can help reduce centralized computation at one site and instead process decentralized data at different sites. Thereby, it is possible to optimize the calculation ability with big data problems such as remote sensing image data. Moreover, by using a semi-supervised technique, it is possible to take advantage of a supervised clustering technique with a small amount of labeled data to improve the accuracy of clustering the results. Experiments on some actual remote sensing image data show that the proposed method gives significantly better results than the previously proposed methods. This result shows the potential of collaborative clustering in big data processing and improves the clustering quality by using semi-supervised fuzzy clustering with a small amount of labeled data. © 2022 IEEE.

关键词： Fuzzy clustering

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：