In remote sensing image-blurring is induced by many sources such asatmospheric scatter, optical aberration, spatial and temporal sensorintegration. The natural blurring can be exploited to speed up target search byfast template matching. In this paper, we synthetically induce additionalnon-uniform blurring to further increase the speed of the matching process. Toavoid loss of accuracy, the amount of synthetic blurring is varied spatiallyover the image according to the underlying content. We extend transitivealgorithm for fast template matching by incorporating controlled image blur. Tothis end we propose an Efficient Group Size (EGS) algorithm which minimizes thenumber of similarity computations for a particular search image. A largerefficient group size guarantees less computations and more speedup. EGSalgorithm is used as a component in our proposed Optimizing auto-correlation(OptA) algorithm. In OptA a search image is iteratively non-uniformly blurredwhile ensuring no accuracy degradation at any image location. In each iterationefficient group size and overall computations are estimated by using theproposed EGS algorithm. The OptA algorithm stops when the number ofcomputations cannot be further decreased without accuracy degradation. Theproposed algorithm is compared with six existing state of the art exhaustiveaccuracy techniques using correlation coefficient as the similarity measure.Experiments on satellite and aerial image datasets demonstrate theeffectiveness of the proposed algorithm.
展开▼
