A Neural Network approach to edge detection using Adaptive Neuro-Fuzzy Inference System

This paper highlights the importance of edge detection in action recognition and presents an edge detection method based on Artificial Neural Network. To implement this concept the Adaptive Neuro-Fuzzy Inference System (ANFIS) has been used. The ANFIS is first designed, trained and checked for average error tolerance. The system is then tested with a few sample images whose results are discussed at the end. A comparison between the traditional edge detectors and the ANFIS method is also provided.