Abstract
A dual stage neural-network architecture is proposed, for the combined local and global processing of visual dot-patterns, in the spatial-orientational space. Dipoles defining local orientation are first determined by a Hopfield-type network. The second stage, consisting of a layered neural network, is then trained by the back propagation algorithm with a new cost function to discriminate between dipole patterns. The network performance in classification tasks is similar to that of human vision.
Original language | English |
---|---|
DOIs | |
State | Published - 1989 |
Externally published | Yes |
Event | 16th Conference of Electrical and Electronics Engineers in Israel, EEIS 1989 - Tel-Aviv, Israel Duration: 7 Mar 1989 → 9 Mar 1989 |
Conference
Conference | 16th Conference of Electrical and Electronics Engineers in Israel, EEIS 1989 |
---|---|
Country/Territory | Israel |
City | Tel-Aviv |
Period | 7/03/89 → 9/03/89 |