arXiv 1809.021472018-09-06cs.CL

升级您的OCR：在罗马化梵文中重用OCR进行OCR后文本校正

Upcycle Your OCR: Reusing OCRs for Post-OCR Text Correction in Romanised Sanskrit

IIT Kharagpur（印度理工学院克勒格布尔分校）；University of California, San Diego（加利福尼亚大学圣迭戈分校）；Walmart Labs, India（沃尔玛实验室（印度））

Amrith Krishna, Bodhisattwa Prasad Majumder, Rajesh Shreedhar Bhat, Pawan Goyal

备注：: This paper has been accepted as a full paper in the SIGNLL Conference on Computational Natural Language Learning (CoNLL), 2018. The code, data and the supplementary material is available at https://github.com/majumderb/sanskrit-ocr

摘要：

我们提出了一个后OCR文本校正的方法，在罗马化梵语的数字化文本。由于缺乏资源，我们的方法使用的OCR模型训练的其他语言写在罗马。目前，没有可用于罗马化梵文OCR的数据集。因此，我们自举了一个包含430张图像的数据集，这些图像在两种不同的设置下扫描，并获得了相应的地面实况。对于训练，我们综合生成两种设置的训练图像。我们发现使用复制机制（Gu et al.，2016）在解决单调序列到序列任务中，字符识别率（CRR）比当前最先进的模型增加了7.69个百分比（Schnober等人，2016年）。我们发现，我们的系统在打击OCR倾向性错误方面是强大的，因为它从OCR输出中获得了87.01%的CRR，其中一个数据集设置的CRR为35.76%。对模型进行的人类判断调查表明，我们提出的模型的预测结果比其他系统更快地理解和改进人类。

英文摘要：

We propose a post-OCR text correction approach for digitising texts in Romanised Sanskrit. Owing to the lack of resources our approach uses OCR models trained for other languages written in Roman. Currently, there exists no dataset available for Romanised Sanskrit OCR. So, we bootstrap a dataset of 430 images, scanned in two different settings and their corresponding ground truth. For training, we synthetically generate training images for both the settings. We find that the use of copying mechanism (Gu et al., 2016) yields a percentage increase of 7.69 in Character Recognition Rate (CRR) than the current state of the art model in solving monotone sequence-to-sequence tasks (Schnober et al., 2016). We find that our system is robust in combating OCR-prone errors, as it obtains a CRR of 87.01% from an OCR output with CRR of 35.76% for one of the dataset settings. A human judgment survey performed on the models shows that our proposed model results in predictions which are faster to comprehend and faster to improve for a human than the other systems.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。