Improving Online Handwriting Text/Non-text Classification Accuracy Under Condition of Stroke Context Absence

Polotskyi, Serhii; Deriuga, Ivan; Ignatova, Tetiana; Melnyk, Volodymyr; Azarov, Hennadii

doi:10.1007/978-3-030-20521-8_18

Cited by 4 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, GRU uses sufficiently fewer cells in the recurrent layers. GRU RNNs are widely used in handwritten text/nontext classifiers [4,7].…”

Section: Proposed Model Architecturementioning

confidence: 99%

“…Handwriting is one of the most natural ways of data input and content creation [1]. It has been widely used in humanmachine interaction [2][3][4][5]. It became possible due to hardware improvement (touchscreen, stylus, CPU, etc.)…”

Section: Introductionmentioning

confidence: 99%

“…There are a lot of publications, focused on binary stroke classification (text/non-text) [3][4][6][7]. But, the multi-class classifiers for the DLA task in online handwritten documents is a new demanding trend that is still in its infancy [2,8].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hierarchical Recurrent Neural Network for Handwritten Strokes Classification

Degtyarenko¹,

Deriuga²,

Grygoriev³

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

The paper presents an original solution to the online handwritten document processing in a free form, which is aimed at separating multi-class handwritten documents into texts, tables, formulas, drawings, etc. Stroke classification is an important step in automatic document layout analysis (DLA) in handwritten document recognition systems. Major DLA challenges arise due to a wide diversity of handwritten content, various writing styles, a lack of contextual knowledge, and the complicated structure of freeform handwritten documents. In this paper, we propose the hierarchical recurrent neural network (RNN) architecture to address the hierarchical structure inherent to the handwritten document. The novelty of feature aggregation pooling technique for transferring data between hierarchical levels allows achieving higher computational efficiency for using the suggested approach in on-device mobile computing. The presented approach gives an access to new state-of-the-art results in the task of multi-class classification with an accuracy of 97.25% on the IAMonDo dataset. This result can serve as the basis for efficient mobile applications for freeform handwriting document recognition.

show abstract

“…In addition, GRU uses sufficiently fewer cells in the recurrent layers. GRU RNNs are widely used in handwritten text/nontext classifiers [4,7].…”

Section: Proposed Model Architecturementioning

confidence: 99%