Title :
Recurrent network for the detection of salient structures in remote sensed images
Author :
Gaio, Mauro ; Madelaine, Jacques
Author_Institution :
LAIAC, Caen Univ., France
fDate :
27 Jun-2 Jul 1994
Abstract :
The paper defines what perceptive phenomenon is behind the human identification of salient linear structures. It then establishes the reasons for using Gestalt laws to find spatial relation factors. The central section presents the recurrent network model and explains the initialization of the cells and the links. Finally an example of the processing on an aerial photography is given and the guidelines of our future research ends the paper
Keywords :
image recognition; recurrent neural nets; remote sensing; Gestalt laws; aerial photography; cell initialization; link initialization; perceptive phenomenon; recurrent neural net; remotely sensed images; salient linear structures; salient structure detection; spatial relation factors; Guidelines; Humans; Image segmentation; Intelligent networks; Photography; Remote sensing; Roads; Signal processing; Solids; Visual perception;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374405
