A model based approach to edge detection and parameterisation

An important area in low level image processing is the problem of accurately detecting edges. These edges can then be used as the basis for further algorithmic analysis, such as counting the number of objects in a scene or matching images for example. In this paper a novel technique is presented for edge detection and parameterisation based on least squares fitting of an edge model to the data. As well as results in edge detection, images restored using the gained parameterisations are shown