Research On Anti-counterfeiting Feature Extraction Algorithm Of Passport Image In Ultraviolet Spectrum

With the deepening of personnel exchanges between our country and other countries,the intelligent authentication of passports has been implemented in the fields of financial logistics,self-service customs clearance,etc.The passport authentication system uses the passport authentication algorithm to detect and judge the multi-spectral images of passports.The focus of the detection is anti-counterfeiting features.Because the anti-counterfeiting features in ultraviolet images are rich and difficult to imitate,ultraviolet anti-counterfeiting features play an important role in the passport authentication system.Due to the unstable quality of the ultraviolet images of passports and the strong interference of the background of passports during the passport authentication process,the authentication result of the ultraviolet anti-counterfeiting features is not ideal.In this thesis,by combining the enhancement of ultraviolet anti-counterfeiting information and the extraction of ultraviolet anti-counterfeiting features,the topic of anti-counterfeiting features extraction of ultraviolet images of passports is studied.Aiming at the problem that the quality of ultraviolet spectral image of passports is unstable and the ultraviolet anti-counterfeiting information is not obvious,we propose an enhancement model of ultraviolet anti-counterfeiting information based on the features of ultraviolet spectrum.On the basis of the existing saliency model and the enhancement model,we improve the detection module of anti-counterfeiting area and the enhancement module of the ultraviolet anti-counterfeiting information.Our model achieves the purpose of improving the enhancement of the ultraviolet anti-counterfeiting information.Aiming at the problem of the interference of the background of passports with the verification results of passports and the difficulty of extracting ultraviolet anticounterfeiting features,we propose an extraction model of ultraviolet anti-counterfeiting features based on superpixel.We improve the extraction model of ultraviolet anticounterfeiting features by introducing a saliency detection model.Our model achieves the purpose of extracting ultraviolet anti-counterfeiting features effectively.Experimental results show that our enhancement model has a significant enhancement effect on ultraviolet anti-counterfeiting information and our extraction model has a significant extraction effect of ultraviolet anti-counterfeiting features,compared with the existing algorithms.Based on our models,the passport authentication algorithm can obtain rich ultraviolet anti-counterfeiting features.