Fast and accurate scene text understanding with image binarization and off-the-shelf OCR

Milyaev, Sergey; Barinova, Olga; Novikova, Tatiana; Kohli, Pushmeet; Lempitsky, Victor

doi:10.1007/s10032-015-0240-4

Cited by 16 publications

(21 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We evaluated several local binarization methods, including Niblack's, Sauvola's, Howe's, etc. [42], and found that Non-linear Niblack's algorithm produced much better results in our case. Non-linear Niblack' algorithm adds two ordered statistics filter to the background and foreground filters respectively, which can effectively handle poor conditions in natural scenes, such as uneven illumination and degraded text.…”

Section: Image Binarizationsupporting

confidence: 49%

Reading car license plates using deep neural networks

Wang

You

et al. 2018

Image and Vision Computing

159

198

View full text Add to dashboard Cite

Section: Image Binarizationsupporting

confidence: 49%

Reading car license plates using deep neural networks

Wang

You

et al. 2018

Image and Vision Computing

159

198

View full text Add to dashboard Cite

“…engine: the open source project Tesseract 6 [62]. Similar pipelines [63,64,65] using off-the-shelf OCR engines have demonstrated state-of-the-art end-toend performance in English-only datasets up to very recently, provided that the text detection module is able to produce good pixel-level segmentation of text.…”

Section: End-to-end Multi-lingual Recognition In Scene Imagesmentioning

confidence: 99%

Improving patch-based scene text script identification with ensembles of conjoined networks

Gómez

Nicolaou

Karatzas

2017

Pattern Recognition

View full text Add to dashboard Cite

show abstract

“…Recently, scene text detection and recognition, two fundamental tasks in the field of computer vision, have become increasingly popular due to their wide applications such as scene text understanding [1], image and video retrieval [2]. Among these applications, extracting Entity of Interest (EoI) is one of the most challenging and practical problems, which needs to identify texts that belong to certain entities.…”

Section: Introductionmentioning

confidence: 99%

EATEN: Entity-Aware Attention for Single Shot Visual Text Extraction

Qin

Liu

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

Extracting entity from images is a crucial part of many OCR applications, such as entity recognition of cards, invoices, and receipts. Most of the existing works employ classical detection and recognition paradigm. This paper proposes an Entity-aware Attention Text Extraction Network called EATEN, which is an end-to-end trainable system to extract the entities without any post-processing. In the proposed framework, each entity is parsed by its corresponding entity-aware decoder, respectively. Moreover, we innovatively introduce a state transition mechanism which further improves the robustness of entity extraction. In consideration of the absence of public benchmarks, we construct a dataset of almost 0.6 million images in three realworld scenarios (train ticket, passport and business card), which is publicly available at https://github.com/beacandler/EATEN. To the best of our knowledge, EATEN is the first single shot method to extract entities from images. Extensive experiments on these benchmarks demonstrate the state-of-the-art performance of EATEN.

show abstract

Fast and accurate scene text understanding with image binarization and off-the-shelf OCR

Cited by 16 publications

References 30 publications

Reading car license plates using deep neural networks

Reading car license plates using deep neural networks

Improving patch-based scene text script identification with ensembles of conjoined networks

EATEN: Entity-Aware Attention for Single Shot Visual Text Extraction

Contact Info

Product

Resources

About