A Currency Recognition App For Visually Impaired People

Abstract

In this paper , we will use computer vision techniques for currency detection , which will run on a computationally low smartphone. the appliance runs on the smarthphone without the necessity for any local server. It's intended for wide and practical use by the blind people. Though we use the INR paper currency as a case in point , our method is unique and can scale to variety of domains beyond the currency. Our solution will use a visible OCR Algorithm detection method. To enable robust detection during a segmented environment, we will firstly crop the bill background by using an step by step evaluated supported IGC(iterative graph cuts) method. We analyze the popularity problem as an object retrieval task, which will run on mobile devices. We analyze the performance on a group of images which are captured in variety of natural environments, which will report an accuracy of 95.6% on 2438 images.