ABSTRACT
The perforated number is an important anti-counterfeiting technique widely used in recent years in security document for protecting document number from being forged. However, forgery detection of perforated number is still made by means of manual inspection, which is not only time-consuming but also error-prone. In this paper, we propose an automatic method for verifying the authenticity of perforated number by analysing the shape of the perforated holes and their alignments in horizontal and vertical direction. First, the image with perforated number is captured by the 10–20× optical zoom, and then the perforated holes are segmented by the maximum between-cluster variance. Second, each perforated hole is labelled by the connected component analysis and then fitted as a circle using the least-squares method for determining the parameters of each hole. Lastly, we propose two criteria to verify the authenticity of the perforated number. One is the new defined roundness measurement called local feature, which is denoted by computing the distance between the contour of the perforated hole and its fitted circle. The other is the measurement of the alignments of all perforated holes in horizontal and vertical direction, which is called global feature. The alignments are determined by means of K-means clustering method. The experimental results show that the proposed method has the correct detection rate over 92%, which means the proposed method is an effective method for recognising the authenticity of the perforated number for security document.
Acknowledgements
This work was supported by the Scientific Research Program of Shanxi Provincial Education Department [Grant No. 15JS076].