Indian Banknote Recognition using Convolutional Neural Network
Shubham Mittal, Shiva Mittal · 2018
This paper presents a deep learning-based method for identification of denominations of Indian Currency Rupee notes from their color images. A classification framework has been implemented using the concept of transfer learning where a large convolutional neural network pre-trained on millions of natural images is employed for classification of images from new classes. An image dataset of four banknote denominations is prepared by preprocessing and augmentation of real-bank note images acquired in different viewpoints and lighting conditions via smartphone camera. A new top layer upon the convolutional base of a pre-trained MobileNet model is trained for a few epochs upon a portion of the dataset to achieve an agreeable accuracy upon validation subset. With no hassle of feature engineering or extensive preprocessing tasks, the retrained lightweight model achieves an accuracy of 96.6 % on a held out testing subset. Experimental results prove it to be employable for development of dedicated portable systems for identification of banknote denominations.