Microfiche was a common format used in microforms reproductions of documents, extensively used for archival storage before the move to digital formats. While contemporary documents are still available for digitization, others from older historical periods are no longer physically accessible for various reasons. In some cases, their microfiche copies are available, making microfiche digitization a must. However, a microfiche reader is not always available and, even then, it is a machine made for the purpose of reading and not for data collection. In this work, the performance two imaging devices are evaluated as alternatives to the traditional microfiche reader, by means of optical character recognition (OCR). Results show that this alternative surpasses the performance of a microfiche reader in terms of text legibility.
We present a novel method for super-resolution (SR) of license plate images based on an end-to-end convolutional neural networks (CNN) combining generative adversial networks (GANs) and optical character recognition (OCR). License plate SR systems play an important role in number of security applications such as improvement of road safety, traffic monitoring or surveillance. The specific task requires not only realistic-looking reconstructed images but it also needs to preserve the text information. Standard CNN SR and GANs fail to accomplish this requirment. The incorporation of the OCR pipeline into the method also allows training of the network without the need of ground truth high resolution data which enables easy training on real data with all the real image degradations including compression.