Neural Network Based Image Edge Detection Within Spiral Architecture
Xiangjian He, Tom Hintz, Qiang Wu · Parallel and Distributed Processing Techniques and Applications · 2002
Edge detection of 3-dimensional physical objects in a 2-dimensional image is one of the main research areas of computer vision. Edge point is defined as a hexagonal pixel where an abrupt change in grey level takes place. In this paper, we present a new approach to edge detection based on a neuml network. Our task is divided into two parts: 1. Mapping typical grey levels in primitive small image blocks (groups of adjacent 7 hexagonal pizels) within the Spirol Arlchitecture. 2. Combining this locally derived information (including presence, orientation and strength of edge) in a consistent way. This new edge detection scheme, because of its parallel structure, is fast and can be easily implemented.