scispace - formally typeset
Search or ask a question
Topic

Document layout analysis

About: Document layout analysis is a research topic. Over the lifetime, 1462 publications have been published within this topic receiving 34021 citations.


Papers
More filters
Proceedings ArticleDOI
25 Aug 2013
TL;DR: A novel multi-modal document indexing framework for retrieval of old and degraded text documents by combining OCR'ed text and image based representation using learning is proposed.
Abstract: The paper proposes a novel multi-modal document image retrieval framework by exploiting the information of text and graphics regions. The framework applies multiple kernel learning based hashing formulation for generation of composite document indexes using different modalities. The existing multimedia management methods for imaged text documents have not addressed the requirement of old and degraded documents. In the subsequent contribution, we propose novel multi-modal document indexing framework for retrieval of old and degraded text documents by combining OCR'ed text and image based representation using learning. The evaluation of proposed concepts is demonstrated on sampled magazine cover pages, and documents of Devanagari script.

5 citations

Posted Content
TL;DR: A new approach to segment and classify the document regions as text, image, drawings and table is proposed and Discipulus tool has been used to construct the Genetic programming based classifier model and located 97.5% classification accuracy.
Abstract: Document segmentation is a method of rending the document into distinct regions. A document is an assortment of information and a standard mode of conveying information to others. Pursuance of data from documents involves ton of human effort, time intense and might severely prohibit the usage of data systems. So, automatic information pursuance from the document has become a big issue. It is been shown that document segmentation will facilitate to beat such problems. This paper proposes a new approach to segment and classify the document regions as text, image, drawings and table. Document image is divided into blocks using Run length smearing rule and features are extracted from every blocks. Discipulus tool has been used to construct the Genetic programming based classifier model and located 97.5% classification accuracy.

5 citations

Patent
03 Jun 1986
TL;DR: In this paper, a titled system is constituted of an input part 1 for inputting data such as images, an output part 2 for displaying or printing out necessary information, an internal storage part 3, an external storage part 4, and a control part for controlling respective parts.
Abstract: PURPOSE: To form a document corresponding to a purpose without using unnecessary labor for the layout of the document by displaying items described by a user in an priority order on the basis of layout structure corresponding to its logical structure. CONSTITUTION: The titled system is constituted of an input part 1 for inputting data such as images, an output part 2 for displaying or printing out necessary information, an internal storage part 3, an external storage part 4, and a control part for controlling respective parts. At the time of formation of a business document, a user inputs only a sentence item to be described and selects layout structure appropriate for the item from the logical structure of the described contents. Then, the described sentence item inputted by the user is converted into layout structure, which is displayed to check whether the use satisfies the layout display or not. When the user does not satisfies the display, similar conversion is applied to the succeeding layout structure and the converted structure is displayed. Thus, the user can select his preferably layout display pattern. COPYRIGHT: (C)1987,JPO&Japio

5 citations

Posted Content
TL;DR: This paper considers scientific document layout analysis as an object detection task over digital images, without any additional text features that need to be added into the network during the training process, to demonstrate that this deep learning architecture is suitable for tasks that lack very large document corpora for training ab initio.
Abstract: We present an approach for adapting convolutional neural networks for object recognition and classification to scientific literature layout detection (SLLD), a shared subtask of several information extraction problems. Scientific publications contain multiple types of information sought by researchers in various disciplines, organized into an abstract, bibliography, and sections documenting related work, experimental methods, and results; however, there is no effective way to extract this information due to their diverse layout. In this paper, we present a novel approach to developing an end-to-end learning framework to segment and classify major regions of a scientific document. We consider scientific document layout analysis as an object detection task over digital images, without any additional text features that need to be added into the network during the training process. Our technical objective is to implement transfer learning via fine-tuning of pre-trained networks and thereby demonstrate that this deep learning architecture is suitable for tasks that lack very large document corpora for training ab initio. As part of the experimental test bed for empirical evaluation of this approach, we created a merged multi-corpus data set for scientific publication layout detection tasks. Our results show good improvement with fine-tuning of a pre-trained base network using this merged data set, compared to the baseline convolutional neural network architecture.

5 citations

Patent
Koji Kurokawa1, Katsuhito Fujimoto1, Misako Suwa1, Yoshinobu Hotta1, Satoshi Naoi1 
25 Jun 2003
TL;DR: In this article, a document information input apparatus detects a position and an attribute of an area of a real document to be input designated by a user with high accuracy, and then pastes the resulting information to a pertinent position of an electronic document on a display.
Abstract: A document information input apparatus detects a position and an attribute of an area of a real document to be input designated by a user with high accuracy. Based on the detected position and attribute, the document information input apparatus recognizes an image of the area as text information by performing recognition processes suitable for the detected attribute such as character recognition, table recognition and a figure process. Then, the document information input apparatus pastes the resulting information to a pertinent position of an electronic document on a display. As a result, it is possible to input information such as a character sequence, a table and a figure from a real document to an electronic document at high speed and with high accuracy.

5 citations


Network Information
Related Topics (5)
Feature extraction
111.8K papers, 2.1M citations
82% related
Feature (computer vision)
128.2K papers, 1.7M citations
82% related
Object detection
46.1K papers, 1.3M citations
81% related
Image segmentation
79.6K papers, 1.8M citations
80% related
Convolutional neural network
74.7K papers, 2M citations
79% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20235
202219
202134
202019
201914
20189