Intelligent texture anti-counterfeiting method based on DCT (Discrete Cosine Transform) transformation

Abstract

The invention discloses an intelligent texture anti-counterfeiting method based on DCT (Discrete Cosine Transform) transformation, belonging to the texture anti-counterfeiting field. The intelligent texture anti-counterfeiting method comprises the following steps of: firstly establishing a textural feature database, to be specific, (1) carrying out full-graph DCT transformation on each original texture label image, and obtaining feature vectors V(n) in a transfer domain, and (2) storing the determined N feature vectors in the textural feature database; and then automatically identifying the images, to be specific, (3) scanning texture label images to be tested by using a mobile phone, determining the visual feature vectors V' of the images to be detected by using the method of the step (1), and uploading the visual feature vectors V' to a server; (4) determining a normalized correlation coefficient NC (N) value between the feature vectors of all the texture images in the feature database and the visual feature vectors V' of the images to be detected, and (5) returning the maximum value of the NC(n) to the mobile phone of a user. Experiments prove that the intelligent texture anti-counterfeiting method has the capacity of automatically identifying the texture images and the network transmission speed is fast.

Description

technical field [0001] The invention relates to an intelligent texture anti-counterfeiting technology based on DCT transformation and image visual features, is a method capable of automatically identifying texture anti-counterfeiting labels, and belongs to the field of texture anti-counterfeiting technology. Background technique [0002] Counterfeit and shoddy goods seriously endanger the legitimate rights and interests of consumers and seriously disrupt the rules of the entire social and economic operation. It is a serious social and political problem. Anti-counterfeiting technology is a technical means used to identify authenticity and prevent counterfeiting and counterfeiting. From the perspective of technical characteristics and functional evolution, current anti-counterfeiting technology can be divided into the following five generations of products: laser labels, query digital anti-counterfeiting labels, texture anti-counterfeiting Labels, security thread anti-counterf...

AI Technical Summary

Problems solved by technology

[0004] The above texture anti-counterfeiting method has the following deficiencies in practical application: 1) manual comparison is required

When comparing texture anti-counterfeiting labels, users must first download clear texture photos from the Internet, which takes a long time

[0005] For this reason, the conventional texture anti-counterfeiting technology has certain shortcomings in terms of intelligence, rapidity and occupied storage space of identification.

In particular, the research on intelligent algorithms for automatic identification has not yet been publicly reported.

Images