2021
DOI: 10.1002/wics.1547
|View full text |Cite
|
Sign up to set email alerts
|

From object detection to text detection and recognition: A brief evolution history of optical character recognition

Abstract: Text detection and recognition, which is also known as optical character recognition (OCR), is an active research area under quick development with a lot of exciting applications. Deep‐learning‐based methods represent the state‐of‐art of this area. However, these methods are largely deterministic: they give a deterministic output for each input. For both statisticians and general users, methods supporting uncertainty inference are of great appeal, leaving rich research opportunities to incorporate statistical … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6
2
2

Relationship

0
10

Authors

Journals

citations
Cited by 16 publications
(5 citation statements)
references
References 115 publications
0
5
0
Order By: Relevance
“…However, integrating automation, big data, and computer-based vision detection systems can signi cantly enhance manufacturing systems [73]. In the last decade, Deep Learning (DL) models have gained recognition due to their ability to power computer vision-related tasks [74,75]. Convolutional Neural Network (CNN) [76] uses a feed-forward topology to propagate signals and is being widely used for image classi cation and object detection [77].…”
Section: Methodsmentioning
confidence: 99%
“…However, integrating automation, big data, and computer-based vision detection systems can signi cantly enhance manufacturing systems [73]. In the last decade, Deep Learning (DL) models have gained recognition due to their ability to power computer vision-related tasks [74,75]. Convolutional Neural Network (CNN) [76] uses a feed-forward topology to propagate signals and is being widely used for image classi cation and object detection [77].…”
Section: Methodsmentioning
confidence: 99%
“…Besides, different cameras capture images with various orientations, resolutions, contrast rations, illuminations, and noises for the same text scene. All these unknown factors may make some of the algorithms fail to detect the text from the input image [13].…”
Section: Text Recognitionmentioning
confidence: 99%
“…In this section, we present a brief review of the most recent seminal deep-learning-based work on Latin text recognition. Text recognition methods can, broadly, fit into a unified segmentation-free framework, consisting of rectification, visual feature extraction, sequence modeling, and transcription [4], [10], [15], [16], [27]- [29]. Some of these methods can handle textlines of arbitrary lengths, while others require resizing to a fixed size.…”
Section: Related Work a Work On Latin Text Recognitionmentioning
confidence: 99%