Top 5 technologies for optical character detection

chinadaily.com.cn
Updated: Dec 4, 2018
Chinese tech company Cloudwalk's algorithm Pixel-Anchor ranked first on list of results on ICDAR 2017 MLT Chinese. [Photo provided to chinadaily.com.cn]

China has grabbed the top five spots in a prestigious global ranking in the optical character recognition field, according to a report by The Paper.

The ranking from the International Conference on Document Analysis and Recognition, an important index in OCR field, was reset by Chinese tech company Cloudwalk's algorithm Pixel-Anchor, which took top spot in the latest single-scale results on ICDAR 2015.

ICDAR2015 is a pure English text detection sub-data set, used to test algorithms of text detection organically through robust reading competitions established by ICDAR in 2003, aiming to improve text detection and recognition level in natural scene.

The robust reading competitions, with open testing data set online and results submitted anytime, have attracted over 3,500 participating teams from 89 countries and regions, including Google, Microsoft, Amazon and Facebook.

Let's take a look at the top 5 technologies in optical character detection.

No 5 Alibaba with IncepText algorith

OCR technology can be applied in various fields such as photo analysis, license plate recognition and image advertisement filtering. [Photo provided to chinadaily.com.cn]

No 4 SenseTime with FOTS algorith

SenseTime with FOTS technology can extract text information of a card image automatically. [Photo/VCG]

No 3 Face++ with Mask Text algorith

Mask Text technology can detect text information from a person's ID card. [Photo/faceid.com]

No 2 China's Nanjing University and Nanjing University of Science and Technology with PSENet algorith

China has grabbed all top five posts of a prestigious global ranking in the optical character recognition field. [Photo provided to chinadaily.com.cn]

No 1 Cloudwalk's algorithm Pixel-Ancho

The ranking of ICDAR was reset by Chinese tech company Cloudwalk's algorithm Pixel-Anchor. [Photo/VCG]