ShabbyPages: A Reproducible Document Denoising and Binarization Dataset
This provides a reproducible dataset for training and benchmarking document processing models, addressing a gap for researchers and practitioners in the field, though it is incremental as it builds on existing augmentation tools.
The authors tackled the lack of large, complex datasets for document denoising and binarization by introducing ShabbyPages, a dataset with over 6,000 clean images and synthetically-noised counterparts, and demonstrated its utility by training convolutional denoisers that remove real noise features with high human-perceptible fidelity.
Document denoising and binarization are fundamental problems in the document processing space, but current datasets are often too small and lack sufficient complexity to effectively train and benchmark modern data-driven machine learning models. To fill this gap, we introduce ShabbyPages, a new document image dataset designed for training and benchmarking document denoisers and binarizers. ShabbyPages contains over 6,000 clean "born digital" images with synthetically-noised counterparts ("shabby pages") that were augmented using the Augraphy document augmentation tool to appear as if they have been printed and faxed, photocopied, or otherwise altered through physical processes. In this paper, we discuss the creation process of ShabbyPages and demonstrate the utility of ShabbyPages by training convolutional denoisers which remove real noise features with a high degree of human-perceptible fidelity, establishing baseline performance for a new ShabbyPages benchmark.