Vision-based 1D barcode localization method for scale and rotation invariant
Inyong Yun, Joongkyu Kim · 2017
In this paper, we use a vision-based image processing technique to propose a method of real-time one-dimensional barcode localization robust for rotation and scale. The proposed method consists of three main steps. The first step generates and analyzes an orientation histogram of input images and removes their background regions and small clutters. The second step analyzes a local entropy-based orientation for the segmentation of a large region that is highly likely to be barcodes. Finally, the third step analyzes and verifies the barcode structures of the selected candidate region. The test results based on the data set available to the public demonstrate the proposed method is more accurate than existing methods by 10% or more.