arrow
Return

Guided Table Structure Recognition Through Anchor Optimization

delete2021-01-01
delete19
delete
OA
AI
K
Khurram Azeem Hashmi *
D
Didier Stricker
M
Marcus Liwicki
M
Muhammad Afzal
M
Muhammad Zeshan Afzal
DOI:10.1109/ACCESS.2021.3103413delete
deleteOriginal
deleteShare
deleteSave
View PDF
Abstract

Abstract

En 中文
This paper presents the novel approach towards table structure recognition by leveraging the guided anchors. The concept differs from current state-of-the-art systems for table structure recognition that naively apply object detection methods. In contrast to prior techniques, first, we estimate the viable anchors for table structure recognition. Subsequently, these anchors are exploited to locate the rows and columns in tabular images. Furthermore, the paper introduces a simple and effective method that improves the results using tabular layouts in realistic scenarios. The proposed method is exhaustively evaluated on the two publicly available datasets of table structure recognition: ICDAR-2013 and TabStructDB. Moreover, we empirically established the validity of our method by implementing it on the previous approaches. We accomplished state-of-the-art results on the ICDAR-2013 dataset with an average F1-measure of 94.19% (92.06% for rows and 96.32% for columns). Thus, a relative error reduction of more than 25% is achieved. Furthermore, our proposed post-processing improves the average F1-measure to 95.46% that results in a relative error reduction of more than 35%. Moreover, we surpassed the baseline results on the TabStructDB dataset with an average F1-measure of 94.57% (94.08% for rows and 95.06% for columns).
Keywords:
Task analysis
Semantics
Portable document format
Object detection
Image recognition
Annotations
Optimization
Deep neural network
Mask R-CNN
document images
object detection
anchor optimization
guided anchors
table structure recognition
table structure extraction
table understanding
AI Summary

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.

Journal

IEEE Access cover
IEEE Access
IF:
3.6
Papers:
9.8W
Citations:
29.4W

Organization

University of Kaiserslautern cover
University of Kaiserslautern
Scholars:
3.9K
Papers: 3.3K
Citations: 4.3K
Cited Papers

Cited Papers

Table structure understanding and its performance evaluation
err2004-07-01
err60
PREAI
errWang, YL; Phillips, IT; Haralick, RM
errShare
errSave
Consideration of coastal carbonate chemistry in understanding biological calcification
err2016-05-07
err0
errOAAI
errAndrea J. Fassbender; Christopher L. Sabine; Kirsten M. Feifel
errShare
errSave
Structure and expression analysis of genes encoding ADP-glucose pyrophosphorylase large subunit in wheat and its relatives
err2016-07-01
err0
errOAAI
errXiao-Wei Zhang; Si-Yu Li; Ling-Ling Zhang; Qiang Yang; Qian-Tao Jiang; Jian Ma; Peng-Fei Qi; Wei Li; Guo-Yue Chen; Xiu-Jin Lan; Mei Deng; Zhen-Xiang Lu; Chunji Liu; Yu-Ming Wei; You-Liang Zheng
errShare
errSave
THE EXCAVATION OF ARCHAIC HOUSES AT AZORIA IN 2005–2006
err2011-01-01
err0
PREAI
errDonald C. Haggis; Margaret S. Mook; Rodney D. Fitzsimons; C. Margaret Scarry; Lynn M. Snyder
errShare
errSave
errShare
errSave
researcher View more