A Computer Vision Approach to Digit Recognition on Pulp Bales

This paper described a computer vision approach for recognizing quality and producer information of pulp bales from digit series stamped on pulp bales. The digit recognition consist of three stages: segmentation of digit series, features extraction, and classification. Segmentation of digit series is based on image thresholding and Randomized Hough Transform. Digit segmentation produces six digit windows. In features extraction two band-pass derivate of Gaussian filters are used, and the resulting gradient field histograms are used after normalization in classification of digits. The digit in the test can be classified 93% correct with a multiple layer perceptron network. Classification results with three other well known classifiers are also reported.

HEIKKONEN Jukka;  MANTYNEN Mika; 

1996-07-30

JRC13430

https://publications.jrc.ec.europa.eu/repository/handle/JRC13430,