The Performance Evaluation Study Of Image Steganalysis Algorithm

As a study hotspot of information security field, Steganography technology has been concerned by many experts and scholars. In the few years steganography has made a lot of progress. With the development of steganography, steganalysis technology appear. The goal of steganalysis is to defeat steganography methods by identifying the presence of hidden information. But the study about steganalysis technology is immature, so the study about Steganalysis arithmetic performance analysis and evaluation system.In this paper, the common steganalysis algorithm performance and indicators analyze, such as Ture Positive Rate, False positive Rate. On the basis, some new indicators are advanced, such as detection capacity and detection complexity of steganalysis algorithm. Using these indicators, we established steganalysis algorithm performance evaluation system. This evaluation system have two functions. First, it can analyze a steganalysis algorithm performance; secondly, it also can give a steganalysis strategy, which is used to choose a best steganalysis algorithm to detect some carrier images. To analyze steganalysis algorithm can result in a better steganalysis algorithm. In this paper, two steganography algorithm are advance. One is an image steganographic based on module operation. This algorithm is based on Module Operation (MO) and Pseudo Random Number(PRN). First, certain binary message is transformed to decimal message, then the quotient and arithmetical compliment of this decimal message, which are embedded to one pixels pair finaly,are gained through module operation.The experiment proved its high performance in security and hiding capacity. Other is a new image steganography based on pseudo random number. The same Pseudo Random Number sequence is generated by Pseudo Random Number Generator (PRNG) in the parameter certain situation. To utilize this nature of PRNG, the Integer Pseudo Random Number(IPRN) is generated and regarded as an auxiliary pixels value. First, two secret information are hided to one IPRN and one pixel value, then the secret information which is in theIPRN is displaced to the pixel value. In this way, the two secret information bits are embedded into the pixelvalue. This method's characteristic is that the capacity of steganography will be enhanced by one times. Theexperiment proved that this method has a very good security performance and the hiding capacity.