G.V. Shagrova, I.N. Topchiev

In this paper new methods of formation and detection of hidden information contained in the latent image, and algorithms for their implementation are presents. In this paper method of forming a digital latent image based on a rasterization not main but hidden image that allows the use of low-resolution images is considered. There was shows the possibility of using digital filtering techniques implemented in the computer systems of mathematics Mathcad and Matlab for detection of hidden information and visualization algorithms of latent image with a help of these systems were developed. Determined that the method for forming a latent image defines the required sequence of standard filters (Matlab and Mathcad), using of which you can identify the latent image. There was investigated the possibility of using wavelet transform for visualization of hidden information and a method for detection of latent image based on wavelet analysis of the latent image. Wavelet functions are laid out the input signal in horizontal, vertical and diagonal detailing coefficients, so a result of these expansions can reveal latent image at the first level of detail. There was determined that the quality of the identified latent image depends of method of forming latent images and the selected wavelet functions. A visual assessment of the quality of hidden information is very low. Therefore, to improve the quality of the identified latent image is needed processing with a help of digital filtering. For the detection of latent images on bank notes, securities, documents developed base combined filter, which is the optimal sequence of filters, depending on the method of forming a latent image. For operational control of latent images on the securities, bank notes and documents developed software which provides a high visual quality of the identified latent image in comparison with known software through the use of combined base filters.