HIGHLIGHTS
- who: Historical Documents and colleagues from the Southwest University, China have published the research: DOI: 10.4018/JDM.322086 *Corresponding Author, in the Journal: (JOURNAL)
- what: The dataset used in the experiment is derived from Yi dataset (described at Sector 3.2) with its annotations, then ATD net and MSER generate two different text-box, then merge algorithm (described at Sector 3.6.1) is used to merge those two different text-box, and generate a new text-box. To solve this problem, this paper proposes a merging algorithm, which combines the T class candidate . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.