Tiny-BDN: An Efficient and Compact Barcode Detection Network

This paper presents a novel approach for accurate barcodes detection in real and challenging environments using compact deep neural networks. Our approach is based on Convolutional Neural Network (<italic>CNN</italic>) and neural network compression, which can detect the four vertexes coordinates of a barcode accurately and quickly. Our approach consists of four stages: (<inline-formula><tex-math notation="LaTeX">$i$</tex-math></inline-formula>) feature extraction by a base network, (<inline-formula><tex-math notation="LaTeX">$ii$</tex-math></inline-formula>) region proposal network (<italic>RPN</italic>) training, (<inline-formula><tex-math notation="LaTeX">$iii$</tex-math></inline-formula>) barcode classification and coordinates regression, and (<inline-formula><tex-math notation="LaTeX">$iv$</tex-math></inline-formula>) weights pruning and recoding. The model is trained in the first three stages and then compressed in the fourth stage to reduce the size of the trained model. In order to remove the effect of geometric distortion during barcode decoding, we add a distortion removal module to the end of the trained model. In experiments, we validate our approach on a challenging large-scale dataset. Compared with previous methods, our method can locate the coordinates of a barcode accurately and quickly and enhance decoding rate through distortion removal. In addition, the storage and memory overheads of our model are reduced through model compression, which shows great potential in industrial applications.