检索结果-内蒙古大学图书馆

SPIE 11th International Symposium on Multispectral image processing and Pattern Recognition (MIPPR) - Remote Sensing image processing, Geographic Information systems, and Other Applications

作者： Fei, Jingjing Wang, Zhicheng Yu, Zhaohui Gu, xi Wei, Gang Tongji Univ Res Ctr CAD Shanghai Peoples R China

ISBN: (纸本)9781510636422

Object detection is a fundamental yet challenging problem in natural scenes and aerial scenes. Although region based deep convolutional neural networks (CNNs) have brought impressive improvements for object detection in natural scenes, detecting oriented objects in aerial images still remains challenging, due to the complexity of the aerial image backgrounds and the large degree of freedom in scale, orientation, and density. To tackle these problems, we propose a novel network, composed of backbone structure with global attention module, multi-scale object proposal network and final oriented object detector, which can efficiently detect small objects, arbitrary direction objects, and dense objects in aerial images. We utilize pyramid pooling blocks as a global attention module on the top of the backbone structure to generate discriminative feature representations, which provide diverse context information and complementary receptive field for the detector. The global attention module can help the model reduce false alarms and incorrect classifications in the complex aerial image backgrounds. The multi-scale object proposal network aims to generate object-like regions at different scales through several intermediate layers. After that, these regions are sent to the detector for refined classification and regression, which can alleviate the problem of variant scales in aerial images. The oriented object detector is designed to generate predictions for inclined box. The quantitative comparison results on the challenging DOTA dataset show that our proposed method is more accurate than baseline algorithms and is effective for objection detection in aerial images. The results demonstrate that the proposed method significantly improves the performance.

关键词： Object detection aerial image convolution neural networks attention mechanism

来源：评论

学校读者我要写书评

暂无评论

SoftPatch: unsupervised anomaly detection with noisy data 22

SoftPatch: unsupervised anomaly detection with noisy data

引用

Proceedings of the 36th International Conference on Neural Information processing systems

作者： xi Jiang Jianlin Liu Jinbao Wang Qian Nie Kai Wu Yong Liu Chengjie Wang Feng Zheng Southern University of Science and Technology Department of Computer Science and Engineering Tencent Youtu Lab

ISBN: (纸本)9781713871088

Although mainstream unsupervised anomaly detection (AD) algorithms perform well in academic datasets, their performance is limited in practical application due to the ideal experimental setting of clean training data. Training with noisy data is an inevitable problem in real-world anomaly detection but is seldom discussed. This paper considers label-level noise in image sensory anomaly detection for the first time. To solve this problem, we proposed a memory-based unsupervised AD method, SoftPatch, which efficiently denoises the data at the patch level. Noise discriminators are utilized to generate outlier scores for patch-level noise elimination before coreset construction. The scores are then stored in the memory bank to soften the anomaly detection boundary. Compared with existing methods, SoftPatch maintains a strong modeling ability of normal data and alleviates the overconfidence problem in coreset. Comprehensive experiments in various noise scenes demonstrate that SoftPatch outperforms the state-of-the-art AD methods on the MVTecAD and BTAD benchmarks and is comparable to those methods under the setting without noise.

关键词：

来源：评论

学校读者我要写书评

暂无评论

VG-Swarm: A Vision-based Gene Regulation Network for UAVs Swarm Behavior Emergence

arXiv

引用

arxiv 2022年

作者： Cai, Yuwei Li, Huanlin Hong, Juncao Xu, Peng Cheng, Hui Zhu, xiaomin Hu, Bingliang Hao, Zhifeng Fan, Zhun Key Lab of Digital Signal and Image Processing of Guangdong Province College of Engineering University of Shantou Guangdong China School of Computer Science and Engineering University of Sun Yat-Sen Guangdong China College of Systems Engineering National University of Defense Technology Hunan Changsha China Xi’an Institute of Optics and Precision Mechanics Shanxi China

Unmanned Aerial Vehicles (UAVs) dynamic encirclement is an emerging field with great potential. Researchers often get inspiration from biological systems, either from macro-world like fish schools or bird flocks etc, or from micro-world like gene regulatory networks (GRN). However, most swarm control algorithms rely on centralized control, global information acquisition, and communications among neighboring agents. In this work, we propose a distributed swarm control method based purely on vision and GRN without any direct communications, in which swarm agents of e.g. UAVs can generate an entrapping pattern to encircle an escaping target of UAV based purely on their installed omnidirectional vision sensors. A finite-state-machine (FSM) describing the behavioral model of each drone is also designed so that a swarm of drones can accomplish searching and entrapping of the target collectively in an integrated way. We verify the effectiveness and efficiency of the proposed method in various simulation and real-world experiments. Copyright © 2022, The Authors. All rights reserved.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

基于模糊规则的插值推理算法综述

引用

计算机学报 2022年第8期45卷 1687-1711页

作者：李方轶李映杨静西北工业大学计算机学院西安710129 空天地一体化大数据技术及应用国家工程实验室西安710129 陕西省语音与图像信息处理重点实验室西安710129

基于模糊规则的近似推理系统是在模糊集和模糊逻辑理论上建立的,在数学、工程、计算机等科学领域得到了迅速的发展.它作为解决建模和推理中不精确和模糊问题的有效工具,利用模糊if-then规则完成推理任务.传统推理机制只能使用稠密的规... 详细信息

基于模糊规则的近似推理系统是在模糊集和模糊逻辑理论上建立的,在数学、工程、计算机等科学领域得到了迅速的发展.它作为解决建模和推理中不精确和模糊问题的有效工具,利用模糊if-then规则完成推理任务.传统推理机制只能使用稠密的规则库进行推理,其能获得有效结论的前提是保证任何输入的观测值都能与现有的模糊规则相匹配.模糊规则插值(Fuzzy Rule Interpolation,FRI)技术能够有效地利用不完备(或稀疏)的规则库实现推理,为没有规则匹配的观测值通过规则插值估计出其结论.本文系统地综述了基于模糊规则的插值推理技术,着重介绍基于α-截集和基于中间规则的两大类插值方法,通过代表性算法揭示了这两类方法的基本思想.同时,还简要介绍了模糊插值推理系统的其他方法和规则插值算法的实际应用.此外,本文汇总了11个常用的评价指标,并对典型算法进行了系统的比较和讨论.最后对未来的研究工作进行了展望.

关键词：模糊规则插值不完备规则库模糊α-截集模糊中间规则模糊插值推理

来源：评论

学校读者我要写书评

暂无评论

A novel image recognition approach using multiscale saliency model and GoogLeNet 18

A novel image recognition approach using multiscale saliency...

引用

18th image processing: algorithms and systems Conference, IPAS 2020

作者： Yang, Guoan Jian, Libo Lu, Zhengzhi Yang, Junjie Liu, Deyang School of Electronic and Information Engineering Xi’an Jiaotong University No.28 Xianning West Road Xi’an CityShaanxi Province710049 China

It is very good to apply the saliency model in the visual selective attention mechanism to the preprocessing process of image recognition. However, the mechanism of visual perception is still unclear, so this visual saliency model is not ideal. To this end, this paper proposes a novel image recognition approach using multiscale saliency model and GoogLeNet. First, a multi-scale convolutional neural network was taken advantage of constructing multiscale salient maps, which could be used as filters. Second, an original image was combined with the salient maps to generate the filtered image, which highlighted the salient regions and suppressed the background in the image. Third, the image recognition task was implemented by adopting the classical GoogLeNet model. In this paper, many experiments were completed by comparing four commonly used evaluation indicators on the standard image database MSRA10K. The experimental results show that the recognition results of the test images based on the proposed method are superior to some state-of-the-art image recognition methods, and are also more approximate to the results of human eye observation. © 2020, Society for Imaging Science and Technology

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

Design of Day-Lily Robot Recognition System Based on ZYNQ

Design of Day-Lily Robot Recognition System Based on ZYNQ

引用

Networking, Communications and Information Technology (NetCIT), International Conference on

作者： xiuxia Zhang Haozhe Ma Shuyi Wei Lailong Yang School of Electronics and Information Engineering North Minzu University Yinchuan China School of Mechanical Engineering Xi'an Jiaotong University Xi'an China

ISBN: (纸本)9781665400718

With the development and wide application of robots in agricultural production, more and more agricultural fruit and vegetable robot schemes had been put forward, but the research on day-lily robot is less. Aiming at the problem of high labor cost in the current day-lily picking process, this paper designed a portable, low-cost embedded recognition system based on ZYNQ based on the characteristics of the ripe fruits of day-lily. A low-cost portable embedded recognition system was designed based on ZYNQ for day-lily robot. ZYNQ processing board was used to recognize the fruit image, and the combination of traditional image processing and machine learning recognition algorithm was used to realize the efficient real-time recognition of the target fruit. Traditional image processing combined with machine learning recognition algorithm could be transplanted to ZYNQ processing board which had the FPGA, which could better guarantee the real-time image recognition. Moreover, this recognition system has high recognition efficiency, suitable accuracy and low cost, which could meet the identification requirements of portable and low cost of agricultural robots at the same time.

关键词： image recognition Costs Machine learning algorithms Target recognition Machine learning Production Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral image Dimension Reduction and Target Detection Based on Weighted Mean Filter and Manifold Learning 3

Hyperspectral Image Dimension Reduction and Target Detection...

引用

2020 3rd International Conference on Computer Information Science and Artificial Intelligence, CISAI 2020

作者： Jiang, Yihe Wang, Tao Chang, Hongwei Su, Yanzhao Xi'An Research Institute of Hi-Tech Xi'an710025 China

Hyperspectral images have many bands, resulting in a high data volume, which complicates subsequent data processing. Using manifold learning methods to reduce the dimension of data is conducive for subsequent use. However, traditional manifold learning methods are easily disturbed by spectral uncertainty, and are primarily used for hyperspectral image classification. This paper proposes an improved manifold reconstruction preserving embedding algorithm based on weighted mean filter (WMF-IMRPE), that is not easily affected by spectral uncertainty and has excellent target detection performance. The original hyperspectral image is processed by the weighted mean filter to eliminate the influence of noise and reduce spectral differences between the homogeneous ground objects. The spectral angular distance then replaces the Euclidean distance in the original MRPE algorithm to select neighbourhood pixels, reducing spectral uncertainty interference. The experimental results show that the low dimensional features extracted by the WMF-IMRPE algorithm have better distinguishability, and the algorithm further improves the target detection accuracy. The WMF-IMRPE algorithm's hyperspectral image target detection performance is superior to other similar algorithms. © Published under licence by IOP Publishing Ltd.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

基于膨胀腐蚀卷积块的单幅图像去雨算法

引用

计算机系统应用 2023年第10期32卷 201-207页

作者：李延恺吴向东吴甜甜长安大学信息工程学院西安710018

成像设备在雨天拍摄图像时由于雨雾和雨条纹的存在会导致图像质量严重退化,对后续图像处理性能造成极大影响.因此,图像的去雨算法研究引起了广泛关注,其中针对单幅图像的去雨算法由于没有先验知识的支持,面临较大挑战.近年来,深度学习... 详细信息

成像设备在雨天拍摄图像时由于雨雾和雨条纹的存在会导致图像质量严重退化,对后续图像处理性能造成极大影响.因此,图像的去雨算法研究引起了广泛关注,其中针对单幅图像的去雨算法由于没有先验知识的支持,面临较大挑战.近年来,深度学习因其高特征表示能力被应用在图像去雨算法研究中.本文基于小波变换,采取了一种深度学习与数字图像形态学处理相结合的算法来实现单幅图像去雨,具有训练参数少、训练时间短和去雨效果好等优点.首先对含雨图像进行小波变换,分为低频分量、水平高频分量、垂直高频分量和对角高频分量,然后对这4个分量分别构造深度学习神经网络,并在神经网络架构中根据雨的特征加入图像膨胀、腐蚀等形态学处理来进行去雨操作,大大简化了模型架构,并能取得较好的结果.

关键词：单幅图像去雨小波变换形态学处理神经网络

来源：评论

学校读者我要写书评

暂无评论

Face matching system in multi-pose changing scene

Face matching system in multi-pose changing scene

引用

6th International Conference on Intelligent Computing and Signal processing (ICSP)

作者： Qianqian Liu Beijing Engineering Research Center for IoT Software and Systems Beijing University of Technology Beijing China

ISBN: (纸本)9781665447300

Due to the emergence of deep learning, face recognition has made remarkable achievements. But in the traditional video surveillance scene, because the monitoring object is not controlled and the face posture changes frequently, the actual face image may be a negative face image, which greatly reduces the accuracy of face recognition. And the traditional monitoring system does not have an automatic warning function, which wastes a lot of human resources. This paper focuses on the realization of a real-time and efficient face matching system in multi-pose changing scenes. The system mainly uses the MTCNN algorithm for face detection and alignment and uses a generative confrontation network to make face positive to eliminate the influence of posture change. The feature extraction network uses lightweight MobileFaceNet as the backbone network for feature extraction, and the feature comparison is carried out by cosine similarity. Experiments show that the multi-pose face matching system can greatly improve the speed of feature extraction, and has good accuracy, real-time, and robustness.

关键词： Training Face recognition Signal processing algorithms Signal processing Feature extraction Video surveillance Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Graph-Order Optimization Algorithm Based on Equal-in-Space Distance Model for High-Resolution image Matting

Graph-Order Optimization Algorithm Based on Equal-in-Space D...

引用

IEEE International Conference on Cloud Computing and Intelligence systems (CCIS)

作者： Fujian Feng Han Huang Yihui Liang School of Software Engineering South China University of Technology Guangzhou China School of Data Science and Information Engineering Guizhou Minzu University Guiyang China Key Laboratory of Big Data and Intelligent Robot South China University of Technology Ministry of Education Guangzhou China School of Computer Science Zhongshan Institute University of Electronic Science and Technology of China Zhongshan China

ISBN: (纸本)9781665441506

image matting is an essential image processing technology. optimized-based image matting methods can significantly improve the alpha matte quality of high-resolution images. However, the local information of the foreground may be similar to the background, which causes the inversion problem of the alpha matte in the single-point optimized. In this paper, we propose an image matting mathematical model of the equal-in-space distance. The model transforms the high-resolution image matting problem into several small-scale combinatorial optimization problems according to the similarity among pixel features. Inspired by spanning tree, we propose a graph-order optimization strategy, which generates the optimization sequence of small-scale optimization problems according to the edge weight among graph nodes. In addition, we designed a graph-order optimization algorithm based on optimized information transfer to solve each node in the graph. Experimental results show that the proposed model solves the alpha matte inversion problem of single-point optimization matting. Besides, the proposed algorithm outperforms the state-of-the-art optimization algorithms for the high-resolution image matting problem.

关键词： Cloud computing image processing image edge detection Conferences Computational modeling Transforms Information processing

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：