Reinforcement Learnig For Parameter Control of Text Detection in Images and Video Sequences

Abstract : A framework for parameterization in computer vision algorithms is evaluated by optimizing ten parameters of the text detection for semantic indexing algorithm preposed by Wolf et al. The Fuzzy ARTMAP neural network is used for generalization, offering much faster learning than in a previous tabular implementation. Difficulties in using a continuous action space are overcome by employing the DIRECT method for global optimization without derivatives. The chosen parameters are evaluated using metrics of recall and precision, and are shown to be superior to the parameters previously recommended.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01593442
Contributor : Équipe Gestionnaire Des Publications Si Liris <>
Submitted on : Tuesday, September 26, 2017 - 11:44:41 AM
Last modification on : Tuesday, February 26, 2019 - 4:35:36 PM

Identifiers

Citation

Graham W. Taylor, Christian Wolf. Reinforcement Learnig For Parameter Control of Text Detection in Images and Video Sequences. International Conference on Information and Communication Technologies : From Theory to Applications, Apr 2004, Damascus, Syria. ⟨10.1109/ICTTA.2004.1307859⟩. ⟨hal-01593442⟩

Share

Metrics

Record views

62