@article{Dr. Ulhas Shiurkar_2021, title={To Detection and Recognition of Marathi Text from Handwritten Document Images using Deep Learning}, url={http://thedesignengineering.com/index.php/DE/article/view/5148}, abstractNote={<p><strong>: </strong>Handwritten character identification from a realistic picture presents many challenges. Bangla characters are made up of a lot of different forms and strokes. Deep learning has recently developed practical skills for extracting high-level features from an image kernel. This article will show how to combine a multilayer convolutional neural network, an origination module, and a fully connected network in a new way. In this system, we proposed a hybrid CNN based handwritten character recognition using deep learning techniques. Various feature extraction and selection techniques have been used during the module training and testing, respectively. In an extensive experimental analysis, we have tested around 5800 image datasets with various cross-validations. The accuracy of proposed system is around 95.20% and evaluated with numerous state-of-arts system<strong>.</strong></p&gt;}, journal={Design Engineering}, author={Dr. Ulhas Shiurkar, Mr. Prashant Sopanrao Kolhe,}, year={2021}, month={Oct.}, pages={2310-2320} }