Enhance to Read Better: A Multi-Task Adversarial Network for Handwritten Document Image Enhancement

26 May 2021  ยท  Sana Khamekhem Jemni, Mohamed Ali Souibgui, Yousri Kessentini, Alicia Fornรฉs ยท

Handwritten document images can be highly affected by degradation for different reasons: Paper ageing, daily-life scenarios (wrinkles, dust, etc.), bad scanning process and so on. These artifacts raise many readability issues for current Handwritten Text Recognition (HTR) algorithms and severely devalue their efficiency. In this paper, we propose an end to end architecture based on Generative Adversarial Networks (GANs) to recover the degraded documents into a clean and readable form. Unlike the most well-known document binarization methods, which try to improve the visual quality of the degraded document, the proposed architecture integrates a handwritten text recognizer that promotes the generated document image to be more readable. To the best of our knowledge, this is the first work to use the text information while binarizing handwritten documents. Extensive experiments conducted on degraded Arabic and Latin handwritten documents demonstrate the usefulness of integrating the recognizer within the GAN architecture, which improves both the visual quality and the readability of the degraded document images. Moreover, we outperform the state of the art in H-DIBCO challenges, after fine tuning our pre-trained model with synthetically degraded Latin handwritten images, on this task.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Binarization DIBCO 2017 A Multi-Task Adversarial Network for Handwritten Document Image Enhancement F-Measure 89.8 # 6
FPS 89.95 # 2
DRD 4.03 # 5
PSNR 17.45 # 6
Binarization H-DIBCO 2012 A Multi-Task Adversarial Network for Handwritten Document Image Enhancement PSNR 22 # 3
F-Measure 95.18 # 3
FPS 94.63 # 1
DRD 1.62 # 3
Binarization H-DIBCO 2016 A Multi-Task Adversarial Network for Handwritten Document Image Enhancement F-Measure 94.95 # 1
FPS 94.55 # 2
PSNR 21.85 # 1
DRD 1.56 # 1
Binarization H-DIBCO 2018 A Multi-Task Adversarial Network for Handwritten Document Image Enhancement PSNR 20.18 # 3
F-Measure 92.41 # 3
FPS 94.35 # 1
DRD 2.6 # 3

Methods


No methods listed for this paper. Add relevant methods here