Indian Paper Currency Recognition Framework for Blind and Visually Impaired People using Deep Learning Model

Indian Paper Currency Recognition Framework for Blind and Visually Impaired People using Deep Learning Model

Publication Date : 2023-08-04
Author(s) :

Anumol R. B, Anuja S. B
Conference Name :

The International Conference on scientific innovations in Science, Technology, and Management (NGCESl-2023)
Abstract :

Currency recognition has always been a troublesome task for blind and visually impaired people (BVIP). Most of the current Indian legal tenders resemble in size, thus making the identification process more difficult. Automated paper currency recognition system, issues such as folded or partial views, uneven illumination, and background clutter make it non-trivial and challenging. This Paper proposes deep learning model for assisting BVIP in recognizing the Indian paper currency denomination. This Paper exploits the latent embeddings of variational auto encoders combined with IPCRNet. Two encoders are trained on currency images to generate two feature maps. The IPCRNet is based on dense connection, Multi-Dilation, and Depth-wise separable convolution layers. Contextual Block (CB) in the backend utilizing the dense connection and dilation scheme in depth-wise separable convolutional layers. The novel contextual part utilizes a depthwise separable convolution for reducing network computations. The multi-dilation scheme offers an enlarged receptive field without increasing the parameters.

No. of Downloads :

18