检索结果-内蒙古大学图书馆

An improved contour detection method based on level set and watershed transform

An improved contour detection method based on level set and ...

MIPPR 2009 - Automatic Target Recognition and image Analysis: 6th International Symposium on Multispectral image processing and Pattern Recognition

作者： Wang, Shuang Fu, Shenggao Jiao, Licheng Zhang, Xiaojing Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9780819478061

In this paper, an improved contour detection method based on level set and watershed transform is proposed. It is performed on coarse-to-fine approach: 1) primary contours detected by using level set evolution;2) accurate contours detected by watershed transform. The combined method utilizes the advantages of level set method and watershed transform. Experimental results show that our method could detect accurate contours and gain great improvement compared with level set evolution and watershed transform. © 2009 Copyright SPIE - The International Society for Optical Engineering.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Improvement of Bandelets in cost function and coding strategy for SAR image compression

Improvement of Bandelets in cost function and coding strateg...

引用

2009 Asia-Pacific Conference on Synthetic Aperture Radar, APSAR 2009

作者： Yang, Shuyuan Wu, Ruixia Meng, Huixiao Jiao, Licheng Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9781424427314

Synthetic aperture radar (SAR) data collections can cover large areas at high resolution, generating massive amounts of data. Many existed transform-based compression techniques can effectively reduce the costs of storage and transmission with slight loss of information. As a new developed multi-resolution geometry analysis tool, Bandelet can make full use of intrinsic geometry regularity of images and exhibit enormous potential in image compression. In this paper, we propose a low complexity Bandelet transform for SAR image compression. A new cost function is defined to determine the optimal geometric flows in each dyadic squared sub-block of images, and an optimization of low frequency component followed by EBCOT coding of the Bandelet coefficients are employed. Some experiments are taken on some Ku-band SAR images came from Sandia National Laboratories and the results show the superiority of our proposed method over JPEG2000 and the second generation Bandelet in PSNR, ENL and EPI. ©2009 IEEE.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Quantum-inspired evolutionary clustering algorithm based on manifold distance 09

Quantum-inspired evolutionary clustering algorithm based on ...

引用

2009 World Summit on Genetic and Evolutionary Computation, 2009 GEC Summit - 1st ACM/SIGEVO Summit on Genetic and Evolutionary Computation, GEC'09

作者： Yangyang, Li Hongzhu, Shi Maoguo, Gong Ronghua, Shang Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9781605583266

Based on the concepts and principles of quantum computing, a quantum-inspired evolutionary algorithm for data clustering (QECA) is proposed in this paper. And a novel distance measurement index called manifold distance is introduced. These attribute data are the main source of clustering problem, due to its complex distribution, most clustering algorithms available are only suitable for these types of characteristic data. In this study, a new algorithm which can deal with these data with manifold distribution is more effective. The main motives of using QECA consist in searching for appropriate cluster center so that a similarity metric of clusters are optimized more quickly and effectively. The superiority of QECA over fuzzy c-means (FCM) algorithm and immune evolutionary clustering algorithm (IECA) is extensively demonstrated in our experiments. Copyright 2009 ACM.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Gene transposon based clonal selection algorithm for clustering

Gene transposon based clonal selection algorithm for cluster...

引用

11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009

作者： Liu, Ruochen Sheng, Zhengchun Jiao, Licheng Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9781605583259

Inspired by the principle of gene transposon proposed by Barbara McClintock, a new immune computing algorithm for clustering multi-class data sets named as Gene Transposition based Clone Selection Algorithm (GTCSA) is proposed in this paper, The proposed algorithm does not require a prior knowledge of the numbers of clustering;an improved variant of the clonal selection algorithm has been used to determine the number of clusters as well as to refine the cluster center. a novel operator called antibody transposon is introduced to the framework of clonal selection algorithm which can realize to find the optimal number of cluster automatically. The proposed method has been extensively compared with Variable-string-length Genetic Algorithm(VGA)based clustering techniques over a test suit of several real life data sets and synthetic data sets. The results of experiments indicate the superiority of the GTCSA over VGA on stability and convergence rate, when clustering multi-class data sets. Copyright 2009 ACM.

关键词： Genes

来源：评论

学校读者我要写书评

暂无评论

Synthetic aperture radar automatic target recognition based on curvelet transform

Synthetic aperture radar automatic target recognition based ...

引用

MIPPR 2009 - Automatic Target Recognition and image Analysis: 6th International Symposium on Multispectral image processing and Pattern Recognition

作者： Wang, Shuang Liu, Zhuo Jiao, Licheng He, Jun Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9780819478061

A novel synthetic aperture radar (SAR) automatic target recognition (ATR) approach based on Curvelet Transform is proposed. However, the existing approaches can not extract the more effective feature. In this paper, our method is concentrated on a new effective representation of the moving and stationary target acquisition and recognition (MSTAR) database to obtain a more accurate target region and reduce feature dimension. Firstly, MSTAR database can be extracted feature through the optimal sparse representation by curvelets to obtain a clear target region. However, considering the loss of part of edges of image. We extract coarse feature, which is to compensate fine feature error brought by segmentation. The final features consisting of fine and coarse feature are classified by SVM with Gaussian radial basis function (RBF) kernel. The experiments show that our proposed algorithm can achieve a better correct classification rate. © 2009 Copyright SPIE - The International Society for Optical Engineering.

关键词： Automatic target recognition

来源：评论

学校读者我要写书评

暂无评论

Algorithm of partition based network boosting for imbalanced data classification

Algorithm of partition based network boosting for imbalanced...

引用

2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010

作者： Shuiping, Gou Hui, Yang Licheng, Jiao Xiong, Zhuang Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education of China Institute of Intelligent Information Processing Xidian University Xi'an 710071 China

ISBN: (纸本)9781424469178

Network Boosting (NB) is an ensemble learning method which combines weak learners together based on a network and can learn the target hypothesis asymptotically. NB has higher generalization ability compared to Bagging and AdaBoost. But, when datasets are class-imbalanced, the performance of NB will decrease quickly. In order to solve this problem, we present a Partition based Network Boosting method (PNB) to classify imbalanced data. For PNB method, every classifier node of the classifier network is provided with the same number of training data which are all of same weights. The classifier in the network is built by the balanced training set sampled from the training data according to the weights record of the training data it holds. And then, the weights of the instances of every node classifier are updated based on the classification results of self-node and its neighbor nodes. The classifier network is trained repeatedly in such a way. Weight factor of hypothesis in the training progress is introduced to improve the performance. The final classification is formed by all the hypotheses of the classifier network learned during the training progress so that the label of new instances can be decided by the weight voting. The experimental results on UCI data and imbalanced biomedical data show that the PNB algorithm has better AUC and recall performance compared with NB learning machine. © 2010 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Zhuang Characteristic Culture Detection Based on Improved YOLOV3-SPP Network 14th

Zhuang Characteristic Culture Detection Based on Improved YO...

引用

14th International Conference on Genetic and Evolutionary Computing, ICGEC 2021

作者： Guo, Shu Qiang, Baohua Xie, Wu Zhai, Yijie Chen, Ruidong Zheng, Hong Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guangxi Guilin541004 China

ISBN: (纸本)9789811684296

In order to enhance the dissemination power of Zhuang culture and enhance the influence of traditional Chinese culture, this article proposes a data set containing the characteristic cultural products of the Zhuang nationality for the unique cultural development of the Zhuang nationality. The data set has a total of 1890 images and contains three categories, namely Zhuang Five-color glutinous rice, Zhuang Bronze drum, and Zhuang Hydrangea. At the same time, the YOLOV3-SPP network based on the darknet53 model is improved, and the ResNet101 structure is introduced to improve the robustness of the network. The verification results on the Zhuang characteristic data set show that the improved YOLOV3-SPP network has an accuracy rate of 92.78% on the Zhuang characteristic data set, which is 1.28% higher than the original YOLOV3-SPP network. The experimental results show that the improved YOLOV3-SPP structure can better realize the detection of the characteristic culture of the Zhuang nationality. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： YOLOV3-SPP model Zhuang characteristic data set Zhuang culture

来源：评论

学校读者我要写书评

暂无评论

Discovering distinctive action parts for action recognition

Discovering distinctive action parts for action recognition

引用

作者： Chen, Feifei Sang, Nong Gao, Changxin Kuang, Xiaoqin Key Laboratory of Ministry of Education for Image Processing and Intelligent Control School of Automation Huazhong University of Science and Technology China

ISBN: (纸本)9781479957514

Recent methods based on mid-level visual concepts have shown promising capability in human action recognition field. Automatically discovering semantic entities such as parts for an action class remains challenging. In this paper, we focus on discovering distinctive action parts for recognition of human actions by learning and selecting a small number of discriminative part detectors directly from training videos. We initially train a large collection of candidate Exemplar-LDA detectors from clusters obtained by clustering spatiotemporal patches in whitened space. A novel Coverage-Entropy curve is proposed as a means of measuring the representative and discriminative capabilities of part detectors, and used to select a set of compact and meaningful detectors out of the vast candidates. By integrating these mined detectors into 'bag of parts' representation, our approach demonstrates state-of-the-art performance on the UCF50 dataset. © 2014 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Real-time Facial Features Tracker Based on Topologic Information and Similarity Analysis

引用

电子学报(英文版) 2008年第2期17卷 205-209页

作者： LIANG Ronghua FANG Jinglong MAO Jianfei ZHOU Delong College of Information Engineering Zhejiang University of Technology Hangzhou 310014 China State Key Laboratory of Audiovisual Information Processing Peking University Beijing 100871 China Institute of Graphics and Image Hangzhou Dianzi Universtity Hangzhou 310018 China

Real-time facial features tracking of video can be widely used in face recognition, video surveillance, face animation and Human-Computer Interaction. We present a fast tracking approach, and our method only requires simple device - a digital camera and a PC, and our approach needs limited user interactions. We first use eigenface and topologic information to detect the position and the size of face from the first frame, and facial features of the first frame are acquired automatically. The successor of the first frame can be tracked by using similarity analysis and motion estimation, the automatic tracker of first frame is also used to resolve the features occlusion problem when the tracked features disappear which is a difficult issue for tracking. Experimental results show that our approach is easily implemented, and the analysis also shows the high robustness of our method.

关键词：拓扑信息相似分析特征跟踪运动评估

来源：评论

学校读者我要写书评

暂无评论

A tube length measurement device based on acoustic method

A tube length measurement device based on acoustic method

引用

2011 IEEE International conference on intelligent Computation and Bio-Medical Instrumentation, ICBMI 2011

作者： Chen, Yimin Qi, Jian Zhang, Xuming Ding, Mingyue Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9780769546230

A low-cost device using acoustic method for measuring open-end tube length is developed. The proposed device is aimed to get the length of the tubes which are piled up together, and only one end of which is available in the warehouse or dumping areas. The device contains the emitting and receiving probes to echo the sound pulse and receive the reflection sound wave, and has been developed with discrete component circuits whose key part is the logical control circuit based on ARM. Experiments are performed in the open tubes with the lengths between 0.85m to 6m. The measurement error is below 0.5cm. For the tubes blocked by the cement or earth, the device can detect the location of obstruction automatically. Based on the acoustic method, experiments are also conducted in the blocked tubes based on Matlab software, and the analysis of the reflection wave is given. © 2011 IEEE.

关键词： MATLAB

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：