Integration of Shape Context and Neural Networks for Symbol Recognition

Abstract : Using shape matching within a k–nearest neighbor approach, shape context descriptor has been applied in several classification problems with outstanding results. However , the application of this frame work on large datasets or online scenarios is challenging due to its computational cost. To over come this limitations, we evaluate the use of shape context as input features for neural networks. We test the proposed method in a problem of recognition of handwritten mathematical symbols. For a total of 75 classes of symbols, we obtained a recognition rate of 89.8%, comparable with a k–nearest neighbor approach, but with reduced time complexity.
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01150830
Contributor : Harold Mouchère <>
Submitted on : Monday, May 11, 2015 - 9:49:04 PM
Last modification on : Wednesday, December 19, 2018 - 3:02:05 PM
Long-term archiving on : Monday, September 14, 2015 - 10:51:53 PM

File

CIFED-19.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : hal-01150830, version 1

Collections

Citation

Frank Julca-Aguilar, Christian Viard-Gaudin, Harold Mouchère, Sofiane Medjkoune, Nina Hirata. Integration of Shape Context and Neural Networks for Symbol Recognition. Colloque International Francophone sur l’Écrit et le Document 2014 (CIFED), Mar 2014, Nancy, France. ⟨hal-01150830⟩

Share

Metrics

Record views

308

Files downloads

98