Distinguishing line-detection from texture segregation using a modular network-based model

In order to enable task-dependent processing of lines, edges, and textures, the authors introduce a network-based model for the front-end visual system. The purpose of this model is to provide different representations of a retinal image in such a way that different actions and decisions about the presence of objects in the visual scene can be undertaken at a further stage. In particular, as a starting point, the model with distinguish lines or edges from textures. A hierarchy of ANN (artificial neural network) modules is introduced for performing two essentially different tasks: line and edge detection and texture segregation. The network module is the Entropy Driven Artificial Neural Network module. The model has been implemented on a parallel computer, a 16-processor MEIKO transputer system. The model´s ability to distinguish line and edge detection from texture segregation, starting from the same bank of Gabor filters, is demonstrated