CHANGE DETECTION AND DYNAMIC ANALYSIS BASED ON REMOTE SENSING IMAGES

Abstract

Abstract. A radar uses the time elapsed between the transmission and reception of an electromagnetic waveform to locate targets present in the illuminated area. Different objects will reflect the radiation with different intensities and phase. The signal provided by standard radar is a profile of the intensity backscattered from the scene as a function of the distance. The resolution, i.e. the capability to distinguish different targets, is related to instrumental parameters and, for conventional radar, is in the range of tens of centimetres. The elementary sampling volume of a radar measurement is usually called radar bin. A radar image can be obtained when an azimuth and a range resolution is available, and this can be attained in different ways: performing a mechanical scanning of the antenna, the most familiar mode used for surveillance, meteorological radar etc, or modifying its spatial features by changing the characteristics of the radiated signal or finally through a specific processing of the acquired data, as in the case of Synthetic Aperture Radar (SAR). In this paper only 1D data without any cross range resolution are used. The vibration of a target corresponds to a small and rapid variation of the radar-target distance to which the phase of the received signal is related. Coherent radar is able to provide measurements of the phase variation along time exploiting the interferometric technique. The received radar signals permits to retrieve distance variations of the observed objects in the order of small fractions of the transmitted wavelength, by comparing the phase of signals acquired at different times. Use a short span bridge as a test-bed this study investigates the actual capability of a Real Aperture Radar (RAR) interferometer to detect the natural vibration caused by wind or pass pedestrians. It is found that RAR can pick up bridge displacements of a few tens of μm and detect a wide range of vibrations.