Journal of Intelligent and Robotic Systems
Change Detection in Overhead Imagery Using Neural Networks
Applied Intelligence
Dynamic detection of nuclear reactor core incident
Signal Processing
| Hi-index | 0.00 |
We propose a multifractal approach to the problem of change detection in image sequences, such as registrated remotely sensed images of the same scene or sequences of medical images. We show that the multifractal analysis of images-based on a modelisation of the two-dimensional signal as measure-can be of great help if we want to detect changes without any a priori knowledge of the objects to be extracted. We first present a simple change detection method based on the classical multifractal analysis of images w.r.t. the Lebesgue measure. We then describe an improved method based on the analysis of images w.r.t. a reference measure, which in this case is the first image of the sequence. We finally show some results on real data.