Abstract
The present work investigates the potential of neural adaptive learning to solve the correspondence problem within a two-frame adaptive area matching approach. A novel method is proposed based on the use of the zero mean normalized cross-correlation coefficient integrated within a neural network model which uses a least-mean-square delta rule for training. Two experiments were conducted for evaluating the neural model proposed. The first aimed to produce dense disparity maps based on the analysis of standard test images. The second experiment, conducted in the biomedical field, aimed to model 3D surfaces from a varied set of scanning electron microscope stereoscopic image pairs.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have