检索结果-内蒙古大学图书馆

2010 20th International Conference on pattern Recognition, ICPR 2010

作者： Chanda, Sukalpa Pal, Umapada Franke, Katrin Kimura, Fumitaka Department of Computer Science and Media Tech. Gjøvik University College Norway Computer Vision and Pattern Recognition Unit Indian Statistical Institute India Graduate School of Engineering Mie University Japan

ISBN: (纸本)9780769541099

All Han-based scripts (Chinese, Japanese, and Korean) possess similar visual characteristics. Hence system development for identification of Chinese, Japanese and Korean scripts from a single document page is quite challenging. It is noted that a Han-based document page might also have Roman script in them. A multi-script OCR system dealing with Chinese, Japanese, Korean, and Roman scripts, demands identification of scripts before execution of respective OCR modules. We propose a system to address this problem using directional features along with a Gaussian Kernel-based Support Vector Machine. We got promising results of 98.39% script identification accuracy at character level and 99.85% at block level, when no rejection was considered. © 2010 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Audio-Visual Model Distillation Using Acoustic Images

Audio-Visual Model Distillation Using Acoustic Images

引用

IEEE Workshop on Applications of computer vision (WACV)

作者： Andrés F. Pérez Valentina Sanguineti Pietro Morerio Vittorio Murino Pattern Analysis & Computer Vision Istituto Italiano di Tecnologia

ISBN: (数字)9781728165530

ISBN: (纸本)9781728165547

In this paper, we investigate how to learn rich and robust feature representations for audio classification from visual data and acoustic images, a novel audio data modality. Former models learn audio representations from raw signals or spectral data acquired by a single microphone, with remarkable results in classification and retrieval. However, such representations are not so robust towards variable environmental sound conditions. We tackle this drawback by exploiting a new multimodal labeled action recognition dataset acquired by a hybrid audio-visual sensor that provides RGB video, raw audio signals, and spatialized acoustic data, also known as acoustic images, where the visual and acoustic images are aligned in space and synchronized in time. Using this richer information, we train audio deep learning models in a teacher-student fashion. In particular, we distill knowledge into audio networks from both visual and acoustic image teachers. Our experiments suggest that the learned representations are more powerful and have better generalization capabilities than the features learned from models trained using just single-microphone audio data.

关键词： Acoustics Visualization Data models Training Microphones Machine learning Synchronization

来源：评论

学校读者我要写书评

暂无评论

CASE-CONTROL DISCRIMINATION THROUGH EFFECTIVE BRAIN CONNECTIVITY

CASE-CONTROL DISCRIMINATION THROUGH EFFECTIVE BRAIN CONNECTI...

引用

International Symposium on Biomedical Imaging

作者： A. Crimi L. Dodero V. Murino D. Sona Pattern Analysis and Computer Vision Istituto Italiano di Tecnologia

ISBN: (纸本)9781509011735

Functional and structural connectivity convey different information about the brain. The integration of these different approaches is receiving growing attention from the research community, as it can shed new light on brain functions. This manuscript proposes a constrained autoregressive model with different lag-orders generating an "effective" connectivity matrix which models the structural connectivity integrating the functional activity. Multiple orders are investigated to observe how different time dependencies influence the effective connectivity. The proposed approach alters an initial structural connectivity representation according to functional data, by minimizing the reconstruction error of an autoregressive model constrained by the structural prior. The model is further validated in a case-control experiment, which aims at differentiating healthy subject and young patients affected by autism spectrum disorder.

关键词： Connectome Effective connectivity Brain imaging DTI fMRI Autoregressive model Case-control Autism Spectrum Disorder ASD

来源：评论

学校读者我要写书评

暂无评论

Complexity optimization of adaptive RBF networks

Complexity optimization of adaptive RBF networks

引用

13th International Conference on pattern Recognition, ICPR 1996

作者： Leonardis, Aleš Bischof, Horst Department for Pattern Recognition and Image Processing Technical University Vienna Treitlstraße 3/1832 A -1040 Vienna Austria University of Ljubljana Faculty of Computer and Information Science Computer Vision Laboratory Ljubljana Slovenia

ISBN: (纸本)081867282X

We propose an extension of RBF networks which includes a mechanism for optimizing the complexity of the network. The approach involves two procedures: adaptation (training) and selection. The first procedure adaptively changes the locations and the width of the centers of the basis functions and trains the linear weights. The selection procedure performs the elimination of some of the basis functions using an objective function. By iteratively combining these two procedures we achieve a controlled way of training and modifying RBF networks, which balances accuracy, learning time, and complexity of the resulting network. © 1996 IEEE.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

Generative Pseudo-label Refinement for Unsupervised Domain Adaptation

Generative Pseudo-label Refinement for Unsupervised Domain A...

引用

IEEE Workshop on Applications of computer vision (WACV)

作者： Pietro Morerio Riccardo Volpi Ruggero Ragonesi Vittorio Murino Pattern Analysis & Computer Vision - Istituto Italiano di Tecnologia

ISBN: (数字)9781728165530

ISBN: (纸本)9781728165547

We investigate and characterize the inherent resilience of conditional Generative Adversarial Networks (cGANs) against noise in their conditioning labels, and exploit this fact in the context of Unsupervised Domain Adaptation (UDA). In UDA, a classifier trained on the labelled source set can be used to infer pseudo-labels on the unlabelled target set. However, this will result in a significant amount of misclassified examples (due to the well-known domain shift issue), which can be interpreted as noise injection in the ground-truth labels for the target set. We show that cGANs are, to some extent, robust against such "shift noise". Indeed, cGANs trained with noisy pseudo-labels, are able to filter such noise and generate cleaner target samples. We exploit this finding in an iterative procedure where a generative model and a classifier are jointly trained: in turn, the generator allows to sample cleaner data from the target distribution, and the classifier allows to associate better labels to target samples, progressively refining target pseudo-labels. Results on common benchmarks show that our method performs better or comparably with the unsupervised domain adaptation state of the art.

关键词： Training Robustness Noise measurement Generative adversarial networks Adaptation models Gallium nitride Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Handwritten bangla compound character recognition using gradient feature

Handwritten bangla compound character recognition using grad...

引用

10th International Conference on Information Technology, ICIT 2007

作者： Pal, U. Wakabayashi, T. Kimura, F. Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata-108 India Graduate School of Engineering Mie University 1577 Kurimamachiya-cho Tsu Mie Japan

ISBN: (纸本)0769530680

Recognition of handwritten characters of Indian script is difficult because of the presence of many complex shaped compound characters (cluster characters) as well as variability involved in the writing style of different individuals. This paper deals with recognition of off-line Bangla handwritten compound characters using Modified Quadratic Discriminant Function (MQDF). The features used for recognition purpose are mainly based on directional information obtained from the arc tangent of the gradient. To get the feature, at first, a 2 X 2 mean filtering is applied 4 times on the gray level image and a non-linear size normalization is done on the image. A Roberts filter is then applied on the normalized image to obtain gradient image. Next, the arc tangent of the gradient (direction of gradient) is initially quantized into 32 directions and the strength of the gradient is accumulated with each of the quantized direction. Finally, the frequencies of these directions are down sampled using Gaussian filter to get 392 dimensional feature vectors. Using 5-fold cross validation technique we obtained 85.90% accuracy from a dataset of Bangla compound characters containing 20,543 samples. © 2007 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A New Defect Detection Method for Improving Text Detection and Recognition Performances in Natural Scene Images

A New Defect Detection Method for Improving Text Detection a...

引用

2020 Swedish Workshop on Data Science, SweDS 2020

作者： Mokayed, Hamam Shivakumara, Palaiahnakote Liwicki, Marcus Pal, Umapada University of Malaya Faculty of Computer Science and Information Technology Kuala Lumpur Malaysia Lulea University of Technology Department of Computer Science Electrical and Space Engineering Sweden Indian Statistical Institute Computer Vision and Pattern Recognition Unit Kolkata India

ISBN: (纸本)9781728192048

This paper presents a new idea for improving text detection and recognition performances by detecting defects in the text detection results. Despite the rapid development of powerful deep learning based models for scene text detection and recognition in the wild, in complex situations (logos or decorated components connected with text), existing methods do not yield satisfactory results. In this paper, we propose to use post-processing method to improve the text detection and recognition performance. The proposed method extracts features, namely phase congruency, entropy and compactness for the text detection results. To strengthen discriminative power for feature extraction, we explore the combination of SVM classifier and Gaussian distribution of text components to determine proper weight, which represents true text component. The weights are multiplied with the features to detect defect components though clustering. The bounding boxes are redrawn, which results proper bounding box without defects components. Experimental results show that the proposed defect detection reports satisfactory results. To validate the effectiveness of defect detection, we conduct experiments on benchmark datasets of MSRA-TD-500 and SVT for detection and recognition before and after defect detection. The result shows that the performance of text detection and recognition improves significantly after defect detection. © 2020 IEEE.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Human motion correction and representation method from motion camera

引用

Journal of Engineering 2017年第1期1卷 370-375页

作者： Zhang, Hong-Bo Guo, Feng Zhang, Miaohui Lin, Ying Hsiao, Tsung-Chih Department of Computer Science and Technology Huaqiao University Xiamen China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Huaqiao University Xiamen China School of Information Science and Engineering Xiamen University Xiamen China Institute of Energy Jiangxi Academy of Sciences Jiangxi Province China

Motion estimation is a basic issue for many computer vision tasks, such as human-computer interaction, motion objection detection and intelligent robot. In many practical scenes, the object movement goes with camera motion. Generally, motion descriptors directly based on optical flow are inaccurate and have low discrimination power. To this end, a novel motion correction method is proposed and a novel motion feature descriptor called the motion difference histogram (MDH) for recognising human action is proposed in this study. Motion estimation results are corrected by background motion estimation and MDH encodes the motion difference between the background and the objects. Experimental results on video shot with camera motion show that the proposed motion correction method is effective and the recognition accuracy of MDH is better than that of the state-of-the-art motion descriptor.

关键词： Motion estimation

来源：评论

学校读者我要写书评

暂无评论

Bangla handwritten character recognition

Bangla handwritten character recognition

引用

2nd Indian International Conference on Artificial Intelligence, IICAI 2005

作者： Roy, K. Pal, U. Kimura, F. Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata-108 India Faculty of Engineering Mie University 1577 Kurimamachiya-cho Mie 514-8504 Japan

ISBN: (纸本)0972741216

Recognition of handwritten characters is difficult because of variability involved in the writing style of different individuals. This paper deals with recognition of off-line Bangla handwritten characters using quadratic classifier based on the features obtained from directional chain code histogram. Here the bounding box of a character is segmented into blocks and chain code histogram is computed in each of the blocks. This chain code features are fed to the quadratic classifier for recognition. We tested the result on 14879 characters and we obtained 90.53% accuracy with 0.81% rejection and 8.66% error, when 50 classes are considered. From 35 classes we obtained 93.90% accuracy with 0.91% rejection and 5.19% error rate. Copyright © IICAI 2005.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

A system for off-line oriya handwritten character recognition using curvature feature

A system for off-line oriya handwritten character recognitio...

引用

10th International Conference on Information Technology, ICIT 2007

ISBN: (纸本)0769530680

In this paper we present a system towards the recognition of off-line Oriya handwritten characters. Since most of the Oriya characters have curve-like stroke, we use curvature feature for the recognition purpose. To get the feature, at first, the input image is size normalized and segmented into 4949 blocks. Curvature is then computed using bi-quadratic interpolation method and quantized into 3 levels according to concave, linear and convex regions. Next direction of gradient is quantized into 32 levels with π/16 intervals, and strength of the gradient is accumulated in each of the 32 directions and in each of the 3 curvature levels of every block. A spatial resolution is made to get 77 blocks from 4949 blocks and a directional resolution is made to get 8 directions from 32 directions. Using curvature features for 3 levels we get 1176 (77 blocks × 8 directions × 3 levels) dimensional features. Finally using principal component analysis we reduce the dimension 1176 to 392 and this 392 dimensional feature vector is fed to a quadratic classifier for recognition. We tested 18190 samples of Oriya handwritten samples and obtained 94.60% accuracy from our proposed system. © 2007 IEEE.

关键词： Principal component analysis

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：