Search

In this paper, dealing with the case of large movements in active vision applications, we first develop an algorithm to estimate the motion of an object and its background. Furthermore, with the assumption of small translations between successive frames, we develop an active tracking algorithm. Its main advantage is that an area-based projection method is presented which resorts to an area integral. Thus, it becomes more robust to the translation distortion of the Log-polar image. In addition, the rotation and scaling estimates can be fulfilled in the spatial domain and not in the frequency domain. Thus, the intrinsic drawbacks of the discrete Fourier transform, such as rotationally dependent aliasing and spectral leakages, can be avoided in our case. Our novelty consists in the introduction of the normalized phase correlation approach in our two algorithms. Because this approach does not rely on the smoothness or differentiability of the flow field in a sequence, it makes the large movement estimation possible. The experimental results show that the motions of object and background can be effectively estimated and a moving object can be tracked using our proposed algorithm in an image sequence.

In this paper, we propose two complete sets of similarity invariant descriptors under the Fourier-Mellin Transform and the Analytical Fourier-Mellin Transform (AFMT) frameworks respectively. Furthermore, their numerical properties are presented and be revealed through image reconstruction. Experimental results indicate that our proposed invariant descriptors can fully reconstruct the original image eliminating any existing similarity transformation (such as rotation, translation and scale) from the original image.

In this paper, the gradient vector flow fields are introduced in the image anisotropic diffusion, and the shock filter, mean curvature flow and Perona-Malik equation are reformulated respectively in the context of this flow fields. Many advantages over the original models can be obtained, such as numerical stability, a large capture range, and computational simplification etc. In addition, the fairing process is introduced in the anisotropic diffusion, which contains the fourth order derivative and is reformulated as the intrinsic Laplacian of curvature under the level set framework. By this fairing process, the boundaries of shape will become more outstanding. In order to overcome numerical errors, the intrinsic Laplacian of curvature is computed from the gradient vector flow fields, but not directly from the observed images.