ON EDGE AND LINE LINKING WITH CONNECTIONIST MODELS

In this paper two connectionist models for mid-level vision problems, namely, edge and line linking, have been presented. The processing ele ments (PE) are arranged in the form of two-dimensional lattice in both the models. The models take the strengths and the corresponding direc tions of the fragmented edges (or lines) as the input. The state of ea ch processing element is updated by the activations received from the neighboring processing elements. In one model, each neuron interacts w ith its eight neighbors, while in the other model, each neuron interac ts over a larger neighborhood. After convergence, the output of the ne urons represent the linked edge (or line) segments in the image. The f irst model directly produces the linked line segments, while the secon d model produces a diffused edge cover. The linked edge segments are f ound by finding out the spine of the diffused edge cover. The experime ntal results and the proof of convergence of the network models have a lso been provided.