Future Work
A future work might come from a development of the currently existing implementation considering the following:
1) Experimental results of the neural network show that the traditional backpropagation learning algorithm may entail the presence of a certain number of weights with a much heigher absolute value than others. Further studies have shown that the main cause of incorrect network functioning on the occurrence of a fault is the non-uniform distribution of absolute values of weights in each layer. So we will try to propose a learning algorithm wich updates the weights, distributing their absolute values as uniformally as possible in each layer.
2) We will extend the concepts of the vertex chain code (VCC) for representing 3D shapes, and use the extended chain code to recognize 3D objects.