A Deep Learning Solution to Detect Text-types using a Convolutional Neural Network
A K M Shahariar Azad Rabby, Md. Majedul Islam, Nazmul Hasan, Jebun Nahar and Fuad Rahman
Accepted to be presented at International Conference on Machine Intelligence & Data Science Application (MIDAS 2020), 4 - 5th September, 2020, Dehradun, India
Description
Text-type detection of a document is an essential pre-processing step in the implementation of many document-processing solutions, such as Optical Character Recognition (OCR) and machine translation. Specifically, text-type detection re-searches for Bangla is very rare, with only a handful of solutions ever reported in the literature. In this paper, we present a lightweight, small footprint convolution-al neural network, which detects handwritten and printed types of content directly from scanned mixed-type document images. The proposed model achieves 99.98% recognition accuracy for this specific two-type classification problem. Text-type detection of a document is an essential pre-processing step in the implementation of many document-processing solutions, such as Optical Character Recognition (OCR) and machine translation. Specifically, text-type detection re-searches for Bangla is very rare, with only a handful of solutions ever reported in the literature. In this paper, we present a lightweight, small footprint convolution-al neural network, which detects handwritten and printed types of content directly from scanned mixed-type document images. The proposed model achieves 99.98% recognition accuracy for this specific two-type classification problem.
