Satellite Image Forgery Detection Based on Buildings Shadows Analysis
Satellite images are to be effectively protected nowadays. There are a lot of ways of changing image content to hide important information: resampling, copy-move, object replacement and other attacks. When these changes are applied to satellite data inclination angles of shadows can be also changed. We propose a new method for satellite image forgery detection based on the analysis of high buildings shadows inclination angles on high resolution snapshots (0.5 m and less). In the proposed solution, the shadows are detected using Canny edge detector with further edge tracing. The comparison of both edge detection methods is presented in the experiments section. The next step is shadows inclination angles estimation using special model-oriented descriptors. The experiments show high accuracy of changed areas detection.
KeywordsSatellite image Forgery High buildings High resolution Shadow detection Inclination angle Canny edge detector Tracing
- 3.Farid, H.: Image forgery detection. IEEE Signal Process. Mag. 16–25 (2009)Google Scholar
- 4.Vladimirovich, K.A., Valerievich, M.V.: A fast plain copy-move detection algorithm based on structural pattern and 2D Rabin-karp rolling hash. In: Campilho, A., Kamel, M. (eds.) ICIAR 2014. LNCS, vol. 8814, pp. 461–468. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-11758-4_50 Google Scholar