Improved QR Code Localization Using Boosted Cascade of Weak Classifiers

Usage of computer-readable visual codes became common in our everyday life at industrial environments and private use. The reading process of visual codes consists of two tasks: localization and data decoding. Unsupervised localization is desirable at industrial setups and for visually impaired people. This paper examines localization efficiency of cascade classifiers using Haar-like features, Local Binary Patterns and Histograms of Oriented Gradients, trained for the finder patterns of QR codes and for the whole code region as well, and proposes improvements in post-processing.