Research On Anti-counterfeit Patterns Of Metal Surface Based On Image Detection And Recognition Algorithm

With the rapid development of computer vision and multimedia, the application ofdigital image processing is more widely. There are many applications in industry,agriculture, medicine, aviation, etc. Metal circulating and commemorative coins are theresearched objects in paper. Through analyzing the features of anti-counterfeit symbolson coins, classical image enhancing and segmenting algorithms are researched, thescheme based on digital image processing for detecting and identifying anti-counterfeitpatterns is proposed. The parsing of anti-counterfeit information is completed.Firstly, in order to extract anti-counterfeit patterns on coins, it needs to researchthe preprocessing algorithms of coin image. It includes image clipping, removingbackground interference, image enhancing, noise filtering, grey level transformation,etc. In enhancement, methods of gray stretching and histogram equalization are used tohighlight images feature, median filtering is used to eliminate or reduce noise, andimage sharpening is used to improve image edge. Secondly, in the anti-counterfeitsegmentation, single algorithm is difficult to deal with the problem in paper. Accordingto different materials coins, segmenting algorithm combined dynamic threshold andedge detecting operator is proposed. Through simulation and comparison, effectivescheme for extracting anti-counterfeit patterns on coins is designed. Through imagemagnification and rotation, connotation of anti-counterfeit information is unscrambledby rearranging patterns. Finally, an improved anti-counterfeit scheme is studied. Theexperiment results show that the schemes in paper for extracting anti-counterfeitpatterns on coins can obtain an ideal effect, and get accurate and clear information.