Use of Neural Networks to Estimate the Number of Nodes of an Edge Quadtree

sion and the resolution desired. The precision and the resolution are defined by one input value, while the shape The number of nodes of an edge quadtree is the measure of its space complexity. This number depends on the figure’s is calculated using the fractal dimension. The proposed shape, its resolution and its precision. The goal of this work is methodology uses the fractal measure also as an input to prove that a relation exists between the number of nodes of value, because this gives more information on how the an edge-quadtree and these three parameters. To reach this image occupies the space. An experimental approximation goal an experimental approach has been used. A unique value of the function is obtained by a suitable training of a neuto represent both the resolution and the precision is used. To ral network. measure the shape of the image we use the fractal dimension. A The second section of this paper describes what an edgemethodology to calculate the fractal dimension and the fractal quadtree is and contains the definition of precision and measure is proposed. These three parameters being given, we resolution. A measure of the precision and of the resolution use a neural network to approximate the sought function. The is given in the third section. In this section some classical computational results show the effectiveness of this approach.  1997 Academic Press methodologies for the definition of the shape are also surveyed, including fractals and box counting techniques. Modifications of these techniques are also considered, in