Filters








17 Hits in 8.2 sec

Can Machine Automatically Discover Text Image from Overall Perspective

Wei Jiang
2019 International Journal of Performability Engineering  
Recently, more and more researchers have focused on the problem about how to automatically distinguish text images from non-text ones.  ...  ; then, random forests are utilized to classify images into text and non-text ones.  ...  Acknowledgement This work was primarily supported by National Natural Science Foundation of China (NSFC) (No.61601184) and Key scientific research project of Education Department of Henan Province (No.16A520018  ... 
doi:10.23940/ijpe.19.01.p28.281287 fatcat:q22kpluwbzbdxga6dbyb4qwsc4

TextField: Learning A Deep Direction Field for Irregular Scene Text Detection [article]

Yongchao Xu, Yukang Wang, Wei Zhou, Yongpan Wang, Zhibo Yang, Xiang Bai
2019 arXiv   accepted
This direction field is represented by an image of two-dimensional vectors and learned via a fully convolutional neural network.  ...  Driven by recent progress in deep learning, impressive performances have been achieved for multi-oriented text detection.  ...  In [33] , the authors propose a Connectionist Text Proposal Network (CTPN) by first predicting vertical text parts, then adopting a recurrent neural network to link text parts.  ... 
arXiv:1812.01393v2 fatcat:npvethdp3bdmtck6bvmfx4mddi

Multi-dimensional long short-term memory networks for artificial Arabic text recognition in news video

Oussama Zayene, Sameh Masmoudi Touj, Jean Hennebert, Rolf Ingold, Najoua Essoukri Ben Amara
2018 IET Computer Vision  
This study presents a novel approach for Arabic video text recognition based on recurrent neural networks.  ...  The proposed system presents a segmentation-free method that relies specifically on a multi-dimensional long short-term memory coupled with a connectionist temporal classification layer.  ...  The multi-dimensional recurrent neural network (MDRNN) architecture [41] represents a generalisation of RNNs, which can deal with multi-dimensional data, e.g. image (2D), video (3D) etc.  ... 
doi:10.1049/iet-cvi.2017.0468 fatcat:qvkmppo5pnaixeujdpllbq4s5e

Learning to Read Irregular Text with Attention Mechanisms

Xiao Yang, Dafang He, Zihan Zhou, Daniel Kifer, C. Lee Giles
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
We present a robust end-to-end neural-based model to attentively recognize text in natural images.  ...  Previous research on text reading often works with regular (horizontal and frontal) text and does not adequately generalize to processing text with perspective distortion or curving effects.  ...  ), then attentively recognize sequence of characters with a recurrent neural network (RNN).  ... 
doi:10.24963/ijcai.2017/458 dblp:conf/ijcai/YangHZKG17 fatcat:xrbs4lz2yvf4jhqdednr4a52oa

Graph-Based Keyword Spotting in Historical Documents Using Context-Aware Hausdorff Edit Distance

Michael Stauffer, Andreas Fischer, Kaspar Riesen
2018 2018 13th IAPR International Workshop on Document Analysis Systems (DAS)  
, text non-text segmentation, page frame segmentation.  ...  The results of an input image after processed by binarization and text non-text segmentation steps without and with our proposed line removal step are shown in Figure 2 .  ... 
doi:10.1109/das.2018.31 dblp:conf/das/Stauffer0R18 fatcat:2r2cjpiitfcs5knjtqbfvcuwsi

Adaptive Boundary Proposal Network for Arbitrary Shape Text Detection [article]

Shi-Xue Zhang, Xiaobin Zhu, Chun Yang, Hongfa Wang, Xu-Cheng Yin
2021 arXiv   pre-print
The adaptive boundary deformation model is an encoder-decoder network, in which the encoder mainly consists of a Graph Convolutional Network (GCN) and a Recurrent Neural Network (RNN).  ...  The boundary proposal model constructed by multi-layer dilated convolutions is adopted to produce prior information (including classification map, distance field, and direction field) and coarse boundary  ...  This work was supported in part by the National Key R&D Program of China (2020AAA09701), National Natural Science Foundation of China (62006018, 62076024).  ... 
arXiv:2107.12664v5 fatcat:jh2qluegybhrxnlueosln6nntu

Few Could Be Better Than All: Feature Sampling and Grouping for Scene Text Detection [article]

Jingqun Tang, Wenqing Zhang, Hongye Liu, MingKun Yang, Bo Jiang, Guanglong Hu, Xiang Bai
2022 arXiv   pre-print
However, these methods cannot well cope with scene text due to its extreme variance of scales and aspect ratios.  ...  Using the basic feature pyramid network for feature extraction, our method consistently achieves state-of-the-art results on several popular datasets for scene text detection.  ...  In an early method, CTPN [45] develops a vertical anchor mechanism to predict sequential proposals, and naturally connects them into bounding boxes by a recurrent neural network.  ... 
arXiv:2203.15221v2 fatcat:ogdex4m5bzgidnn7d2jaya3xni

Towards End-to-End Text Spotting in Natural Scenes [article]

Peng Wang, Hui Li, Chunhua Shen
2021 arXiv   pre-print
In this work, we propose a unified network that simultaneously localizes and recognizes text with a single forward pass, avoiding intermediate processes such as image cropping and feature re-calculation  ...  Text spotting in natural scene images is of great importance for many image understanding tasks. It includes two sub-tasks: text detection and recognition.  ...  Recurrent Neural Networks (RNNs) are employed for this purpose. The work in [41] and [6] formulates word recognition as one-dimensional sequence labeling problem using RNNs.  ... 
arXiv:1906.06013v6 fatcat:6yijskgdkjd2tposdbgw7xtfvq

A Text Detection Algorithm for Image of Student Exercises Based on CTPN and Enhanced YOLOv3

Langcai Cao, Hongwei Li, Rongbiao Xie, Jinrong Zhu
2020 IEEE Access  
It performs two pixel-wise predictions: text/non-text prediction and link prediction.  ...  However, once the characters in the image are very close, separating them using only text/non-text semantic segmentation becomes extremely difficult. Tian et al.  ... 
doi:10.1109/access.2020.3025221 fatcat:b62mxoyjcbf5tknfywdekzyklq

Intelligent Character Recognition System Using Convolutional Neural Network

S. Suriya, Dhivya S, Balaji M
2020 EAI Endorsed Transactions on Cloud Systems  
Convolutional Neural Network differs from other approaches by extracting the features automatically.  ...  Intellectual Character Recognition System is an application that uses Convolutional Neural Network (CNN) to recognize the Tamil character dataset accurately developed by HP Labs India.  ...  high dimensional images.  ... 
doi:10.4108/eai.16-10-2020.166659 fatcat:rrv3tyk2ezegdhcwsvuvvkgbrq

Text Recognition in the Wild: A Survey [article]

Xiaoxue Chen, Lianwen Jin, Yuanzhi Zhu, Canjie Luo, Tianwei Wang
2020 arXiv   pre-print
Therefore, text recognition in natural scenes has been an active research field in computer vision and pattern recognition.  ...  In recent years, with the rise and development of deep learning, numerous methods have shown promising in terms of innovation, practicality, and efficiency.  ...  Recent works [196] , [81] used a convolution neural network (CNN) to improve text/non-text discrimination. • Text detection: The function of text detection [219] , [201] is to determine whether text  ... 
arXiv:2005.03492v3 fatcat:rmzmavxylnf6rbp52lje2mrgiy

A Pipeline Approach to Context-Aware Handwritten Text Recognition

Yee Fan Tan, Tee Connie, Michael Kah Ong Goh, Andrew Beng Jin Teoh
2022 Applied Sciences  
After that, the text sequences are fed to a Residual Network with a Transformer (ResNet-101T) model to perform transcription.  ...  The proposed model is comprised of an object detection neural network that extracts text sequences present on the page regardless of size, orientation, and type (handwritten text, printed text, or non-text  ...  The author proposed the integration of the attention mechanism with the Multi-dimensional Long Short-Term Memory Recurrent Neural Network (MDLSTM-RNN) for implicit text segmentation and transcription.  ... 
doi:10.3390/app12041870 fatcat:fsq3bztcqnaizbpy2obwjeglye

iDocChip: A Configurable Hardware Accelerator for an End-to-End Historical Document Image Processing

Menbere Kina Tekleyohannes, Vladimir Rybalkin, Muhammad Mohsin Ghaffar, Javier Alejandro Varela, Norbert Wehn, Andreas Dengel
2021 Journal of Imaging  
Optical character recognition (OCR) is typically applied to scanned historical archives to transcribe them from document images into machine-readable texts.  ...  We demonstrate our results on multiple platforms with respect to runtime and power consumption.  ...  [64] proposed a novel architecture of a convolutional neural network called MSP-Net for text/non-text image classification.  ... 
doi:10.3390/jimaging7090175 pmid:34564101 pmcid:PMC8467298 fatcat:jsgsgmvfmrg3flzpkpeuoyvq2q

A texture-based pixel labeling approach for historical books

Maroua Mehri, Petra Gomez-Krämer, Pierre Héroux, Alain Boucher, Rémy Mullot
2015 Pattern Analysis and Applications  
One important challenge is to refine well-known approaches based on strong a priori knowledge (e.g. the document image content, layout, typography, font size and type, scanning resolution, image size,  ...  It does not assume a priori information regarding document image content and structure.  ...  -Neural network-based methods (e.g.  ... 
doi:10.1007/s10044-015-0451-9 fatcat:b5njvrjrzjf75iq4t6lvzbqu2m

Text-detection and -recognition from natural images

Hanaa Mahmood
2020
algorithms of the text/non-text regions.  ...  The aim of the study was to develop a robust text detection and recognition method from natural images with high accuracy and recall, which would be used as the target of the experiments.  ...  Previous research has attempted to use filters to solve the problem of text/non-text region classification.  ... 
doi:10.26174/thesis.lboro.11816487 fatcat:u5m5vrca6bbyzb5jlrlyqaqcoy
« Previous Showing results 1 — 15 out of 17 results