检索结果-内蒙古大学图书馆

recognition of handprinted bangla numerals using neural network models 5th

5th International Conference on Asian Fuzzy Systems Society, AFSS 2002

作者： Bhattacharya, U. Das, T.K. Datta, A. Parui, S.K. Chaudhuri, B.B. Computer Vision and Pattern Recognition Unit Indian Statistical Institute 203 B. T. Road Calcutta35 India Computer and Statistical Service Center Indian Statistical Institute 203 B. T. Road Calcutta35 India

ISBN: (纸本)9783540431503

This paper proposes an automatic recognition scheme for hand printed Bangla (an Indian script) numerals using neural network models. A Topology Adaptive Self Organizing Neural Network is first used to extract from a numeral pattern a skeletal shape that is represented as a graph. Certain features like loops, junctions etc. present in the graph are considered to classify a numeral into a smaller group. If the group is a singleton, the recognition is done. Otherwise, multilayer perceptron networks are used to classify different numerals uniquely. The system is trained using a sample data set of 1880 numerals and we obtained 90.56% correct recognition rate on a test set of another 3440 samples. The proposed scheme is sufficiently robust with respect to considerable object noise. © Springer-Verlag Berlin Heidelberg 2002.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

recognition of Multi-oriented Touching Characters in Graphical Documents

Recognition of Multi-oriented Touching Characters in Graphic...

引用

Indian Conference on computer vision, Graphics & Image Processing (ICVGIP)

作者： Partha Pratim Roy Umapada Pal Josep Lladós Computer Vision Center Universitat Autònoma Barcelona Bellaterra Spain Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India

Touching characters are major problem of achieving higher recognition rate in optical character recognition (OCR). Present OCR systems do not perform well when adjacent characters touch. If characters are touched in graphical documents (e.g. map) then such touching string recognition is more difficult because in such documents touching characters appear in multi-oriented direction. In this paper, we present a scheme towards the recognition of English two-character multi-oriented touching strings. When two or more characters touch, they generate a big cavity region at the background portion and we used this background information in our scheme. To handle the background information, convex hull is used. In this scheme, at first, a set of initial segmentation points is predicted based on the concave residues of the convex hull of the touching characters. Next, based on the initial points, we select some candidate segmentation lines. Finally the recognition confidence of two sub-images of a touching string, obtained from each candidate segmentation line is computed. The candidate segmentation line from which we get optimum confidence is the actual segmentation line and the corresponding characters in favour of which the two segmentation parts show optimum confidence is the recognition result of the touching string. To compute the recognition confidence, SVM classifier is used. The features used in the SVM are invariant to character orientation. Circular ring and convex hull ring based approach has been used along with angular information of the contour pixels of the character to make the feature rotation invariant. From the experiment we obtained encouraging result.

关键词： Character recognition Optical character recognition software pattern recognition Support vector machines Support vector machine classification Image segmentation Frequency Image recognition Character generation Degradation

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Unconditioned Deep Generative Synthesis of Retinal Images 20th

Evaluation of Unconditioned Deep Generative Synthesis of Ret...

引用

20th International Conference on Advanced Concepts for Intelligent vision Systems, ACIVS 2020

作者： Kaplan, Sinan Lensu, Lasse Laaksonen, Lauri Uusitalo, Hannu Computer Vision and Pattern Recognition Laboratory Lappeenranta-Lahti University of Technology LUT P.O. Box 20 Lappeenranta53850 Finland Department of Ophthalmology Faculty of Health and Biotechnology Tampere University and Tays Eye Center Tampere Finland

ISBN: (纸本)9783030406042

Retinal images have been increasingly important in clinical diagnostics of several eye and systemic diseases. To help the medical doctors in this work, automatic and semi-automatic diagnosis methods can be used to increase the efficiency of diagnostic and follow-up processes, as well as enable wider disease screening programs. However, the training of advanced machine learning methods for improved retinal image analysis typically requires large and representative retinal image data sets. Even when large data sets of retinal images are available, the occurrence of different medical conditions is unbalanced in them. Hence, there is a need to enrich the existing data sets by data augmentation and introducing noise that is essential to build robust and reliable machine learning models. One way to overcome these shortcomings relies on generative models for synthesizing images. To study the limits of retinal image synthesis, this paper focuses on the deep generative models including a generative adversarial network and a variational autoencoder to synthesize images from noise without conditioning on any information regarding to the retina. The models are trained with the Kaggle EyePACS retinal image set, and for quantifying the image quality in a no-reference manner, the generated images are compared with the retinal images of the DiaRetDB1 database using common similarity metrics. © 2020, Springer Nature Switzerland AG.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

DocSegTr: An Instance-Level End-to-End Document Image Segmentation Transformer

arXiv

引用

arXiv 2022年

作者： Biswas, Sanket Banerjee, Ayan Lladós, Josep Pal, Umapada Computer Vision Center Computer Science Department Universitat Autònoma de Barcelona Spain Computer Vision and Pattern Recognition Unit Indian Statistical Institute India

Understanding documents with rich layouts is an essential step towards information extraction. Business intelligence processes often require the extraction of useful semantic content from documents at a large scale for subsequent decision-making tasks. In this context, instance-level segmentation of different document objects (title, sections, figures etc.) has emerged as an interesting problem for the document analysis and understanding community. To advance the research in this direction, we present a transformer-based model called DocSegTr for end-to-end instance segmentation of complex layouts in document images. The method adapts a twin attention module, for semantic reasoning, which helps to become highly computationally efficient compared with the state-of-the-art. To the best of our knowledge, this is the first work on transformer-based document segmentation. Extensive experimentation on competitive benchmarks like PubLayNet, PRIMA, Historical Japanese (HJ) and TableBank demonstrate that our model achieved comparable or better segmentation performance than the existing state-of-the-art approaches with the average precision of 89.4, 40.3, 83.4 and 93.3. This simple and flexible framework could serve as a promising baseline for instance-level recognition tasks in document images. © 2022, CC BY-SA.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Motion constraint patterns

Motion constraint patterns

引用

1993 IEEE Workshop on Qualitative vision, WQV 1993

作者： Fermüller, Cornelia Computer Vision Laboratory Center for Automation Research University of Maryland College ParkMD20742-3275 United States Department for Pattern Recognition and Image Processing Institute for Automation Technical University Vienna Treitlstraße 3 ViennaA-1040 Austria

ISBN: (纸本)0818636920

The problem of egomotion recovery has been treated by using as input local image motion, with the published algorithms utilizing the geometric constraint relating 2-D local image motion (optical flow, correspondence, derivatives of the image flow) to 3-D motion and structure. Since it has proved very difficult to achieve accurate input (local image motion), a lot of effort has been devoted to the development of robust techniques. A new approach to the problem of egomotion estimation is taken, based on constraints of a global nature. It is proved that local normal flow measurements form global patterns in the image plane. The position of these patterns is related to the three dimensional motion parameters. By locating some of these patterns, which depend only on subsets of the motion parameters, through a simple search technique, the 3-D motion parameters can be found. The proposed algorithmic procedure is very robust, since it is not affected by small perturbations in the normal flow measurements. As a matter of fact, since only the sign of the normal flow measurement is employed, the direction of translation and the axis of rotation can be estimated with up to 100% error in the image measurements. © 1993 IEEE.

关键词： Flow measurement

来源：评论

学校读者我要写书评

暂无评论

SVM Based Scheme for Thai and English Script Identification

SVM Based Scheme for Thai and English Script Identification

引用

International Conference on Document Analysis and recognition

作者： S. Chanda O.R. Terrades U. Pal Indian Statistical Institute Computer Vision and Pattern Recognition Unit Kolkata India Computer Vision Center Universitat Authnòma de Barcelona Barcelona Spain

In some Thai documents, a single text line of a document page may contain both Thai and English scripts. For the optical character recognition (OCR) of such a document page it is better to identify, at first, Thai and English script portions and then to use individual OCR system of the respective scripts on these identified portions. In this paper, a SVM based method is proposed for identification of word-wise printed English and Thai scripts from a single line of a document page. Here, at first, the document is segmented into lines and then lines are segmented into character groups (words). In the proposed scheme, we identify the script of the individual character group combining different character features obtained from structural shape, profile, component overlapping information, topological properties, water reservoir concept etc. Based on the experiment on 6110 data we obtained 99.36% script identification accuracy from the proposed scheme.

关键词： Support vector machines Optical character recognition software Support vector machine classification computer vision Optical network units Structural shapes Water resources Reservoirs Neural networks pattern recognition

来源：评论

学校读者我要写书评

暂无评论

A system for Indian postal automation

A system for Indian postal automation

引用

International Conference on Document Analysis and recognition

作者： K. Roy S. Vajda U. Pal B.B. Chaudhuri A. Belaid Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India LORIA Research Center Nancy France

In this paper, we present a system towards Indian postal automation based on the recognition of pin-code and city name of the postal document. In the proposed system, at first, non-text blocks (postal stamp, postal seal etc.) are detected and destination address block (DAB) is identified from the document. Next, lines and words of the DAB are segmented. Since India is a multi-lingual and multi-script country, the address part may be written by combination of two scripts. To identify the script by which a word is written, we propose a water reservoir based technique. It is very difficult to identify the script by which the pin-code portion is written. So, we have used two-stage artificial neural network (NN) based general classifiers for the recognition of pin-code digits written in English/Bangla. For recognition of city names, we propose an NSHP-HMM (non-symmetric half plane-hidden Markov model) based technique.

关键词： Automation Cities and towns pattern recognition Seals Water resources Reservoirs Neural networks Natural languages Data mining Image recognition

来源：评论

学校读者我要写书评

暂无评论

Diving into the Depths of Spotting Text in Multi-Domain Noisy Scenes

arXiv

引用

arXiv 2023年

作者： Das, Alloy Biswas, Sanket Pal, Umapada Lladós, Josep Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India The Computer Vision Center Computer Science Department Universitat Autónoma de Barcelona Barcelona Spain

When used in a real-world noisy environment, the capacity to generalize to multiple domains is essential for any autonomous scene text spotting system. However, existing state-of-the-art methods employ pretraining and fine-tuning strategies on natural scene datasets, which do not exploit the feature interaction across other complex domains. In this work, we explore and investigate the problem of domain-agnostic scene text spotting, i.e., training a model on multi-domain source data such that it can directly generalize to target domains rather than being specialized for a specific domain or scenario. In this regard, we present the community a text spotting validation benchmark called Under-Water Text (UWT) for noisy underwater scenes to establish an important case study. Moreover, we also design an efficient super-resolution based end-to-end transformer baseline called DA-TextSpotter which achieves comparable or superior performance over existing text spotting architectures for both regular and arbitrary-shaped scene text spotting benchmarks in terms of both accuracy and model efficiency. The dataset, code and pre-trained models will be released upon acceptance. © 2023, CC BY-SA.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Diving into the Depths of Spotting Text in Multi-Domain Noisy Scenes

Diving into the Depths of Spotting Text in Multi-Domain Nois...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Alloy Das Sanket Biswas Umapada Pal Josep Lladós Computer Vision and Pattern Recognition Unit Indian Statistical Institute Kolkata India Computer Science Department Computer Vision Center Universitat Autónoma de Barcelona Barcelona Spain

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

关键词： Training Codes Accuracy Superresolution Benchmark testing Transformers Data models

来源：评论

学校读者我要写书评

暂无评论

Video Script Identification Using a Combination of Textural Features

Video Script Identification Using a Combination of Textural ...

引用

International IEEE Conference on Signal-Image Technologies and Internet-Based System

作者： Zumra Malik Ali Mirza Akram Bennour Imran Siddiqi Chawki Djeddi Center of Computer Vision & Pattern Recognition Bahria University Islamabad Pakistan Departement of Mathematics and Computer Science Larbi Tebessi University Tebessa Algeria

This paper presents a system for script recognition of the text appearing in video frames. The textual content in videos is generally extracted and recognized for development of text based indexing and retrieval systems. If the text in videos appears only in a single script, the output of text detector is directly fed to a video Optical Character recognition (OCR) system for recognition. However, in cases where text may appear in multiple scripts, a script recognition module is required to recognize the script of the text so that it can be processed by the respective OCR. We propose a video script recognition system that considers text in each script as a unique texture. A number of texture measures are extracted from text blocks and an artificial neural network is trained to learn to distinguish between different scripts. The system evaluated on video text blocks in five different scripts (Arabic, English, Urdu, Hindi and Chinese) reported promising recognition rates. In addition to the performance of individual textural features, different combinations of texture measures were investigated which realized interesting results.

关键词： Text recognition Image recognition Histograms Optical character recognition software Fractals Character recognition Indexing

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：