paper-with-me

홈 › Papers

A Study of Augmentation Methods for Handwritten Stenography Recognition

2023-03-05 · Raphaela Heil, Eva Breznik

One of the factors limiting the performance of handwritten text recognition (HTR) for stenography is the small amount of annotated training data. To alleviate the problem of data scarcity, modern HTR methods often employ data augmentation. However, due to specifics of the stenographic script, such settings may not be directly applicable for stenography recognition. In this work, we study 22 classical augmentation techniques, most of which are commonly used for HTR of other scripts, such as Latin handwriting. Through extensive experiments, we identify a group of augmentations, including for example contained ranges of random rotation, shifts and scaling, that are beneficial to the use case of stenography recognition. Furthermore, a number of augmentation approaches, leading to a decrease in recognition performance, are identified. Our results are supported by statistical hypothesis testing. Links to the publicly available dataset and codebase are provided.

📄 PDF Abstract BibTeX arXiv:2303.02761

Code (0)

등록된 구현이 없습니다.

Tasks

Data AugmentationHandwritten Text RecognitionHTR

Similar Papers 제목 키워드 기반

Handwritten Stenography Recognition and the LION Dataset

2023-08-15 · Raphaela Heil, Malin Nauwerck

Purpose: In this paper, we establish a baseline for handwritten stenography recognition, using the novel LION dataset, and investigate the impact of including selected aspects of stenographic theory into the recognition …

Handwritten Text Recognition

Handwritten image augmentation

2023-08-26 · Mahendran N

In this paper, we introduce Handwritten augmentation, a new data augmentation for handwritten character images. This method focuses on augmenting handwritten image data by altering the shape of input characters in traini…

Data AugmentationImage AugmentationOptical Character RecognitionPosition

A Comparative Analysis on Bangla Handwritten Digit Recognition with Data Augmentation and Non-Augmentation Process

2020-06-26 · International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA) 2020 6 · MD Abdullah Al Nasim, Refat E Ferdous, Mahim Anzum Haque Pantho, Atiqul Islam Chowdhury

Determination of Bangla handwritten digit is a momentous image classification task. Though object recognition technology is getting smarter day by day, still Bangla handwritten digit recognition remains inconclusive. …

Data AugmentationHandwritten Digit Recognitionimage-classificationImage Classification+1

StackMix and Blot Augmentations for Handwritten Text Recognition

2021-08-26 · Alex Shonenkov, Denis Karachev, Maxim Novopoltsev, Mark Potanin 외

This paper proposes a handwritten text recognition(HTR) system that outperforms current state-of-the-artmethods. The comparison was carried out on three of themost frequently used in HTR task datasets, namely Ben-tham, I…

Data AugmentationHandwritten Text RecognitionHTRText Generation

Uncovering the Handwritten Text in the Margins: End-to-end Handwritten Text Detection and Recognition

2023-03-10 · Liang Cheng, Jonas Frankemölle, Adam Axelsson, Ekta Vats

The pressing need for digitization of historical documents has led to a strong interest in designing computerised image processing methods for automatic handwritten text recognition. However, not much attention has been …

Data AugmentationHandwritten Text RecognitionText DetectionTransfer Learning