For different document automation operations it is always needed to have an OCR evaluation phase to select the most interesting OCRs for the document class studied. The evaluation should indicate the defects and drawbacks of each OCR and allow to determine the required heuristics to combine these OCRs in order to obtain the highest performances in production: the lowest reject rate for a predef...

Decorated characters are widely used in various documents. Practical optical character reader is required to deal with not only common fonts but also complex designed fonts. However, since appearances of decorated characters are complicated, most general character recognition systems cannot give good performances on decorated characters. In this paper, an algorithm that can extract character’s ...

Neume musical notation is a type of writing of Christian Orthodox Church chant originated in Ancient Byzantium, which is still used by the Orthodox Church. The big varieties of preserved historical documents, which contain neumes give reach material for investigations in fields like history, cultural sciences, etc. This causes the natural need of a software package which can help these investig...

Thai typed character recognition has been a very most popular research topic in Thailand. There are three commercial Thai OCR softwares available to the public at the present. But none of them can preserve the type styles of the original document image such as bold, italics, normal, and bold & italics styles into the output text file. This paper presents the technique for preserving the specifi...

The ocr protein, the product of gene 0.3 of bacteriophage T7, is a structural mimic of the phosphate backbone of B-form DNA. In total it mimics 22 phosphate groups over ~24 bp of DNA. This mimicry allows it to block DNA binding by type I DNA restriction enzymes and to inhibit these enzymes. We have determined that multiple ocr dimers can bind stoichiometrically to the archetypal type I enzyme, ...

We demonstrate that state-of-the-art optical character recognition (OCR) based on deep learning is vulnerable to adversarial images. Minor modifications to images of printed text, which do not change the meaning of the text to a human reader, cause the OCR system to “recognize” a different text where certain words chosen by the adversary are replaced by their semantic opposites. This completely...

Optical character recognition refers to the process of translating images of handwritten, typewritten, or printed text into a format understood by machines for the purpose of editing, indexing/searching, and a reduction in storage size. The OCR process is complicated by noisy inputs, image distortion, and differences between typefaces, sizes, and fonts. Artificial neural networks are commonly u...

Text image super-resolution is a challenging yet open research problem in the computer vision community. In particular, low-resolution images hamper the performance of typical optical character recognition (OCR) systems. In this article, we summarize our entry to the ICDAR2015 Competition on Text Image Super-Resolution. Experiments are based on the provided ICDAR2015 TextSR dataset [3] and the ...

Text categorization in the form of topic identification is a capability of current interest. This paper is concerned with categorization of electronic document images. Previous work on the categorization of document images has relied on Optical Character Recognition (OCR) to provide the transformation between the image domain and a domain where pattern recognition techniques are more readily ap...

Most current commercial optical character recognition (OCR) systems can accurately recognize the text in documents written in a single language. However, when dealing with Greek characters embedded in predominantly English text, these systems do not perform well, and most OCR systems do not recognize the characters as belonging to the Greek alphabet. As a result, the degree of manual review req...