Research And Implementation Of Digital Image Copying And Pasting Blind Forensics System Based On SIFT

With the rapid development of multimedia technology,digital image acquisition devices,including digital cameras,are becoming more and more popular.Digital image is an information carrier that plays an important role in people's daily life and the Internet.At the same time,image processing technology is becoming more and more mature,and with image processing software,people can easily process and edit images.Although these image processing techniques can bring fun to people's lives,they also give some people with ulterior motives a chance.They arbitrarily tamper with the image and confuse it to achieve ulterior motives.The tampering with the abuse of images has caused harmful effects that cannot be ignored.Therefore,the technology to identify the authenticity of images is particularly important,and has become an important topic of current research.Through the research and analysis of image passive forensics algorithms,Qt and OpenCV are used to design a complete set of image copy and paste.The research work of the blind forensic system is mainly:(1)For the existing SIFT-based image copying and pasting tampering detection method,the matching speed is not fast enough and the matching precision is low.An effective feature matching method n2NN(new 2 Nearest Neighbourhood)is proposed.n2 NN is based on The improvement made by the 2NN matching strategy,the method needs to recursively divide the feature point set into two halves before performing 2NN feature matching,and then perform 2NN feature matching on each component semi-set.In this paper,the SIFT-based image copy-and-paste blind forensic method is adopted.Firstly,the SIFT features are extracted,then n2 NN is used to match the feature points,and then the obtained matching pairs are clustered.According to the clustering result,the imitation between the copied block and the pasted block is obtained.Shot transformation relationship model.Finally,the ZNCC method and the morphological operation method are used to quickly locate the tamper region.It can be seen from the experimental results that the improved algorithm is greatly improved in efficiency,and a better tampering area positioning effect is obtained and is still effective for multiple copying and pasting.(2)According to the actual needs of the image forensics system,a complete image copy and paste blind forensic system was designed and implemented.The whole system is divided into four modules,which are read and save modules,display modules,tamper detection modules and add digital watermark modules.At the end of the paper,the system was tested for performance and tested on multiple tamper images.The experimental results confirm that the system designed in this paper can achieve good detection results and has certain application value in the field of image forensics.