A Gaussian Process Upsampling Model for Improvements in Optical Character Recognition
This addresses unreliable image quality in automated document evaluation for financial contexts, but appears incremental.
The paper tackled the problem of low-resolution document images in OCR by applying a Gaussian Process upsampling model, resulting in improved OCR and extraction performance.
Optical Character Recognition and extraction is a key tool in the automatic evaluation of documents in a financial context. However, the image data provided to automated systems can have unreliable quality, and can be inherently low-resolution or downsampled and compressed by a transmitting program. In this paper, we illustrate the efficacy of a Gaussian Process upsampling model for the purposes of improving OCR and extraction through upsampling low resolution documents.