Overview

Original Research

SEGMENTATION OF NON-TEXT FROM BILINGUAL REAL-TIME OFFICE DOCUMENT IMAGES USING U-NET ARCHITECTURE

SHIVAKUMAR G 1, RAVIKUMAR M 2, SAMPATHKUMAR S 3, and SHIVAPRASAD B J 4.

Vol 17, No 07 ( 2022 )   |  DOI: 10.5281/zenodo.6910993   |   Author Affiliation: Department of Computer Science, Kuvempu University, Jnanasahyadri, Shivamogga, India 1,2,3; Department of Computer Science and Engineering, Srinivasa Institute of Technology, Mangalore, India 4.   |   Licensing: CC 4.0   |   Pg no: 811-827   |   To cite: SHIVAKUMAR G, et al., (2022). SEGMENTATION OF NON-TEXT FROM BILINGUAL REAL-TIME OFFICE DOCUMENT IMAGES USING U-NET ARCHITECTURE. 17(07), 811–827. https://doi.org/10.5281/zenodo.6910993   |   Published on: 21-07-2022

Abstract

In this work, we have presented an efficient approach for segmentation of non-text document information from real time office document images which are bilingual using a machine learning approach i.e., U-net architecture for experimentation purpose. We have created our own dataset containing 200 document images. Initially pre-processing is applied on the input document images proposed method is compared with other existing methods and obtained accuracy of 99% different performance measure i.e., (Specificity, Sensitivity, Precision, F1-Score) used in the experimentation.

Keywords

Document Images; Pre-Processing; Filtering; Segmentation (U-Net).