Scene Text Recognition That Eliminates Background and Character Noise Interference

In natural photographs, complex background noise and character noise frequently interfere with scene text identification. To solve the aforementioned concerns, this paper proposes a novel scene character identification model that eliminates noise from both the backdrop and the character (ENBC). The...

Full description

Saved in:
Bibliographic Details
Main Authors: Shancheng Tang, Yaoqian Cao, Shaojun Liang, Zicheng Jin, Kun Lai
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/7/3545
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849730980587241472
author Shancheng Tang
Yaoqian Cao
Shaojun Liang
Zicheng Jin
Kun Lai
author_facet Shancheng Tang
Yaoqian Cao
Shaojun Liang
Zicheng Jin
Kun Lai
author_sort Shancheng Tang
collection DOAJ
description In natural photographs, complex background noise and character noise frequently interfere with scene text identification. To solve the aforementioned concerns, this paper proposes a novel scene character identification model that eliminates noise from both the backdrop and the character (ENBC). The model is divided into three pieces. To begin, the high-level character feature extraction module uses ASPP dilated convolution with varying expansion rates to obtain features at various scales, thereby expanding the receptive field to capture the character feature area more effectively, eliminating noise interference from the character itself, and improving the character shape features. Second, the multi-level character feature fusion module merged the high-level character feature information after upsampling with the low-level character feature information in the backbone network, separated the foreground characters from background interference, removed background noise, and output the resulting image. Third, the recognition enhancement module enhances character context modeling by considering both forward (left-to-right) and backward (right-to-left) information from the text sequence. The experimental results show that the model can effectively minimize background and character noise interference, boosting recognition accuracy by at least 4.2% on the synthetic scene dataset. When compared to other popular techniques on the IIIT5K, ICDAR-2015, ICDAR-2003, and CUTE80 public datasets, recognition accuracy improves by an average of 6.97%.
format Article
id doaj-art-5d0da7e49dbb4f12ae4c486fd46d59d4
institution DOAJ
issn 2076-3417
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-5d0da7e49dbb4f12ae4c486fd46d59d42025-08-20T03:08:43ZengMDPI AGApplied Sciences2076-34172025-03-01157354510.3390/app15073545Scene Text Recognition That Eliminates Background and Character Noise InterferenceShancheng Tang0Yaoqian Cao1Shaojun Liang2Zicheng Jin3Kun Lai4College of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710054, ChinaCollege of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710054, ChinaCollege of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710054, ChinaCollege of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710054, ChinaCollege of Communication and Information Engineering, Xi’an University of Science and Technology, Xi’an 710054, ChinaIn natural photographs, complex background noise and character noise frequently interfere with scene text identification. To solve the aforementioned concerns, this paper proposes a novel scene character identification model that eliminates noise from both the backdrop and the character (ENBC). The model is divided into three pieces. To begin, the high-level character feature extraction module uses ASPP dilated convolution with varying expansion rates to obtain features at various scales, thereby expanding the receptive field to capture the character feature area more effectively, eliminating noise interference from the character itself, and improving the character shape features. Second, the multi-level character feature fusion module merged the high-level character feature information after upsampling with the low-level character feature information in the backbone network, separated the foreground characters from background interference, removed background noise, and output the resulting image. Third, the recognition enhancement module enhances character context modeling by considering both forward (left-to-right) and backward (right-to-left) information from the text sequence. The experimental results show that the model can effectively minimize background and character noise interference, boosting recognition accuracy by at least 4.2% on the synthetic scene dataset. When compared to other popular techniques on the IIIT5K, ICDAR-2015, ICDAR-2003, and CUTE80 public datasets, recognition accuracy improves by an average of 6.97%.https://www.mdpi.com/2076-3417/15/7/3545scene text recognitiondeep learningbackground noisecharacter’s own noiseenhanced character features
spellingShingle Shancheng Tang
Yaoqian Cao
Shaojun Liang
Zicheng Jin
Kun Lai
Scene Text Recognition That Eliminates Background and Character Noise Interference
Applied Sciences
scene text recognition
deep learning
background noise
character’s own noise
enhanced character features
title Scene Text Recognition That Eliminates Background and Character Noise Interference
title_full Scene Text Recognition That Eliminates Background and Character Noise Interference
title_fullStr Scene Text Recognition That Eliminates Background and Character Noise Interference
title_full_unstemmed Scene Text Recognition That Eliminates Background and Character Noise Interference
title_short Scene Text Recognition That Eliminates Background and Character Noise Interference
title_sort scene text recognition that eliminates background and character noise interference
topic scene text recognition
deep learning
background noise
character’s own noise
enhanced character features
url https://www.mdpi.com/2076-3417/15/7/3545
work_keys_str_mv AT shanchengtang scenetextrecognitionthateliminatesbackgroundandcharacternoiseinterference
AT yaoqiancao scenetextrecognitionthateliminatesbackgroundandcharacternoiseinterference
AT shaojunliang scenetextrecognitionthateliminatesbackgroundandcharacternoiseinterference
AT zichengjin scenetextrecognitionthateliminatesbackgroundandcharacternoiseinterference
AT kunlai scenetextrecognitionthateliminatesbackgroundandcharacternoiseinterference