method: edge_link_it42018-08-09

Authors: jin ma

Description: edge_link_it4

method: StradVision2015-03-23

Authors: Hojin Cho et al.

Description: First, we extract character candidates using extremal regions (ER) Second, we verify the extracted character candidates with the character classifier trained by Agile Learning. Afterwards, we do text-patch matching which greatly enhances the recall rate, and group the characters into text regions. Finally, we apply a deep neural network for character recognition.

method: BUCT_YST2015-01-12

Authors: Wei Hu

Description: Multiple MSERs and Convolution Neural Networks are employed in our method to extract character candidates. Character candidates are then grouped by using the technique in USTB_TexStar to generate text candidates. At last, a text classifier is trained on the training dataset to verify these text candidates. The related demo of the software has been released on http://research.cs.buct.edu.cn/huwei ( or http://124.205.208.198:8081).

Ranking Table

Description Paper Source Code
DateMethodPx. RecallPx. PrecisionPx. F-scoreWell s.MergedBrokenBr.-Mer.LostFalse p.DetectedRecallPrecisionFscore
2018-08-09edge_link_it486.85%87.51%87.18%477221411087191553081.32%86.17%83.67%
2015-03-23StradVision78.80%89.24%83.70%4417316120112384507675.27%86.96%80.69%
2015-01-12BUCT_YST74.56%81.75%77.99%42583093451262201499872.56%84.37%78.02%
2013-04-09I2R_NUS_FAR74.73%81.70%78.06%4080317811462355499869.53%81.49%75.04%
2017-02-18TexFESBv1160.32%79.93%68.76%39062401421706259465666.56%83.78%74.19%
2013-04-08NSTextractor60.71%76.28%67.61%3757108701996345445064.03%84.36%72.80%
2013-04-07USTB_FuStar69.58%74.45%71.93%40913111801448966550969.72%74.12%71.85%
2013-04-08I2R_NUS73.57%79.04%76.21%3590743701528357462061.18%77.66%68.44%
2013-04-08NSTsegmentator68.41%63.95%66.10%404615618016482792734168.95%54.98%61.18%
2013-04-05OTCYMIST46.11%58.53%51.58%254736134029264549775943.40%32.62%37.25%
2018-04-19ER_flow_res32_frm5886.70%87.72%87.21%482422514180411723082.21%0.43%0.86%
2018-04-16scene text segmentation and detecion with resNet3287.60%88.40%88.00%481624911079211423082.07%0.43%0.86%

Ranking Graphic

Ranking Graphic