DUET: Detection Utilizing Enhancement for Text in Scanned or Captured Documents

TL;DR

This paper introduces DUET, a deep neural model that combines text detection with enhancement tasks to improve accuracy in noisy scanned document images, utilizing synthetic data and a two-phase training process.

Contribution

The paper proposes a novel multi-task learning approach with auxiliary enhancement tasks and a two-phase training scheme using synthetic and real data for improved document text detection.

Findings

Outperforms existing text detection methods on real document datasets.

Synthetic data and auxiliary enhancement tasks significantly improve detection accuracy.

Two-phase training effectively leverages synthetic and real data for robust performance.

Abstract

We present a novel deep neural model for text detection in document images. For robust text detection in noisy scanned documents, the advantages of multi-task learning are adopted by adding an auxiliary task of text enhancement. Namely, our proposed model is designed to perform noise reduction and text region enhancement as well as text detection. Moreover, we enrich the training data for the model with synthesized document images that are fully labeled for text detection and enhancement, thus overcome the insufficiency of labeled document image data. For the effective exploitation of the synthetic and real data, the training process is separated in two phases. The first phase is training only synthetic data in a fully-supervised manner. Then real data with only detection labels are added in the second phase. The enhancement task for the real data is weakly-supervised with information from their detection labels. Our methods are demonstrated in a real document dataset with performances exceeding those of other text detection methods. Moreover, ablations are conducted and the results confirm the effectiveness of the synthetic data, auxiliary task, and weak-supervision. Whereas the existing text detection studies mostly focus on the text in scenes, our proposed method is optimized to the applications for the text in scanned documents.