Improvement of image binarization methods using image preprocessing with local entropy filtering for alphanumerical character recognition purposes

H Michalak, K Okarma - entropy, 2019 - mdpi.com
Automatic text recognition from the natural images acquired in uncontrolled lighting
conditions is a challenging task due to the presence of shadows hindering the shape …

[PDF][PDF] Fast Binarization of Unevenly Illuminated Document Images Based on Background Estimation for Optical Character Recognition Purposes.

H Michalak, K Okarma - J. Univers. Comput. Sci., 2019 - researchgate.net
One of the key operations during the image preprocessing step in Optical Character
Recognition (OCR) algorithms is image binarization. Although for uniformly illuminated …

Region based adaptive binarization for optical character recognition purposes

H Michalak, K Okarma - 2018 International Interdisciplinary PhD …, 2018 - ieeexplore.ieee.org
Optical Character Recognition (OCR) applications usually require the use of uniformly
illuminated images which can be obtained using the flatbed scanners. However, rapid …

Fast adaptive image binarization using the region based approach

H Michalak, K Okarma - Artificial Intelligence and Algorithms in Intelligent …, 2019 - Springer
Adaptive binarization of unevenly lightened images is one of the key issues in document
image analysis and further text recognition. As the global thresholding leads to improper …

Binarization of degraded document images with generalized Gaussian distribution

R Krupiński, P Lech, M Tecław, K Okarma - Computational Science–ICCS …, 2019 - Springer
One of the most crucial steps of preprocessing of document images subjected to further text
recognition is their binarization, which influences significantly obtained OCR results. Since …

[PDF][PDF] Binarization of document images using the modified local-global Otsu and Kapur algorithms

P Lech, K Okarma, D Wojnar - Przeglad Elektrotechniczny, 2015 - pe.org.pl
In the paper an algorithm for binarization of grayscale images representing text documents
together with its optimization is presented. In the proposed approach two classical global …

Research of the Defects in PET Preform

A Laucka, D Andriukaitis… - 2016 21st International …, 2016 - ieeexplore.ieee.org
Image technologies nowadays are used not only for kee** personal events safe, but also
are widely applied in conjunction with automated electronic systems. Computer vision is …

Fast histogram based image binarization using the Monte Carlo threshold estimation

P Lech, K Okarma - … Vision and Graphics: International Conference, ICCVG …, 2014 - Springer
In the paper the idea of universal fast image binarization method is discussed which utilizes
the histogram estimation using the Monte Carlo approach. Proposed reduction of the …

Fast statistical image binarization of colour images for the recognition of the QR codes

K Okarma, P Lech - Elektronika Ir Elektrotechnika, 2015 - eejournal.ktu.lt
The article concerns the fast image binarization based on the application of the statistical
Monte Carlo method applied for the recognition of the QR codes from colour images …

Improved two-step binarization of degraded document images based on Gaussian mixture model

R Krupiński, P Lech, K Okarma - … , Amsterdam, The Netherlands, June 3–5 …, 2020 - Springer
Image binarization is one of the most relevant preprocessing operations influencing the
results of further image analysis conducted for many purposes. During this step a significant …