Synthetic Arabic Data for Scene Text Recognition

Synthetic Arabic Data for Scene Text Recognition

The dataset consists of 50,000 cropped images with embedded Arabic text. The labels were generated from an Arabic words corpus, which consists of 15 thousand words. The second dataset was collected from Twitter Arabic hashtags and contins 100 cropped images. The datasets were used in our publuished...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: NIDDAL IMAM
Format: Dataset
Sprache:eng
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The dataset consists of 50,000 cropped images with embedded Arabic text. The labels were generated from an Arabic words corpus, which consists of 15 thousand words. The second dataset was collected from Twitter Arabic hashtags and contins 100 cropped images. The datasets were used in our publuished paper "Detecting Spam Images with Embedded Arabic Text in Twitter".
DOI:10.17632/gfc32vndz8