Edge detection using neural network arbitration

Edge detection is a common starting step in many image processing applications. Many varied edge detection techniques have been proposed. Different edge detectors present distinct and different responses to the same image, showing different detail. Our proposal is that different edge detection techniques be utilised in parallel, allowing for the attainment of several edge maps containing different features. The merging of the different maps is performed by a neural network arbitrator. To overcome the increased computing time, implementation is being performed on a multi-transputer array, using the inherent parallelism of the techniques involved. We present examples that show the performance of the neural arbitration of different techniques, along with a comparison with the individual techniques alone