As is shown on ICDAR Robust Reading 2015 official website, on October 17, Dahua Technology took the first place in Task [Word Recognition] of Incidental Scene Text Challenge and Born-Digital Image Challenge; with an accuracy of 82.76pc and 97.43pc respectively.
The International Conference on Document Analysis and Recognition (ICDAR) is an international academic conference, every two years in a different city; a forum for researchers in document analysis. The ICDAR Robust Reading Competition has been held five times; in 2003, 2005, 2011, 2013 and 2015.
Born-Digital Image is one of two challenges dating from ICDAR 2011. It refers to images saved by digital devices from the internet and emails. Automatically extracting text from born-digital images is an interesting prospect as it would provide the enabling technology for a number of applications such as improved indexing and retrieval of web content, enhanced content accessibility, content filtering (such as advertisements or spam emails). And Incidental Scene Text, a new challenge at the 2015 edition in France, refers to text that appears in a scene, before you can improve its positioning or quality in the frame; with applications in wearable cameras and urban scenes generally.
Robust Reading
“Robust Reading” refers to the research area dealing with the interpretation of written communication in unconstrained settings. It has significant implication to video surveillance applications such as vehicle license plate recognition (LPR), container serial number recognition, logistics label context recognition and the recognition of text captured in normal surveillance.
Dahua OCR
Dahua’s AI OCR team from Dahua Advanced Research Institute took part in the ICDAR Robust Reading 2015 competition. Based upon deep learning and the SENet and ResNet network structure, Dahua team developed a way to get accuracy of result. The technologies used in this competition have been applied in Dahua’s smart transportation solutions, such as dealing with tilted car plates, with recognition rate reaching up to 99.99 per cent.
As the company says, deep-learning has led to breakthroughs in video analytics. The accuracy of recognition can be better than human beings; making it possible and economically viable to automate many tasks not possible before. AI has been already applied in public security, transport and banking to protect property and people.
