TNCR: Table Net Detection and Classification Dataset

TL;DR

This paper introduces TNCR, a comprehensive dataset for table detection and classification in document images, along with strong baseline results using deep learning methods to advance research in this area.

Contribution

The paper provides a new, publicly available dataset with diverse images and benchmarks state-of-the-art deep learning models for table detection and classification.

Findings

Cascade Mask R-CNN achieved 79.7% precision

Recall was 89.8%, and F1 score was 84.4%

TNCR dataset is open source and ready for research use

Abstract

We present TNCR, a new table dataset with varying image quality collected from free websites. The TNCR dataset can be used for table detection in scanned document images and their classification into 5 different classes. TNCR contains 9428 high-quality labeled images. In this paper, we have implemented state-of-the-art deep learning-based methods for table detection to create several strong baselines. Cascade Mask R-CNN with ResNeXt-101-64x4d Backbone Network achieves the highest performance compared to other methods with a precision of 79.7%, recall of 89.8%, and f1 score of 84.4% on the TNCR dataset. We have made TNCR open source in the hope of encouraging more deep learning approaches to table detection, classification, and structure recognition. The dataset and trained model checkpoints are available at https://github.com/abdoelsayed2016/TNCR_Dataset.