A Bound On Matching Error

摘要

In matching of the images that are obtained from different sourdces one often observes inherent discrepancies that exist among them. These discrepancies can caue degradation in the object recognition performance, or erros in image registration system that are the basis of some automatic navigational systems. In this paper, a lower bound on the matching error is devived by using a model that treats the two images as being the two ends of a communication channel. Then rate distortion theory is applied to obtain a relationship between the matching error and the statistical properties of the images. The resulting bound can be used to predict the matching erorr, or used in the selection of the proper model and reflerence images, and in the performance evaluation of the target recognition systems.