The upper and lower bounds of the information-hiding capacity of digital images

The information-hiding capacity of a digital image is the maximum information that can be hidden in that image, while the lower limit of information hiding is the minimum detectable information capacity. This paper proposes a new method of information-hiding capacity bounds analysis that is based on the neural network theories of attractors and attraction basins. With this method, the processes for determining upper and lower limits of information hiding, are unified within a single theoretical framework. Results of research show that the theory of attraction basins of neural networks can be used to determine the upper limit of information hiding and the theory of attractors of neural networks can be used to determine the lower limit of information hiding.