Shape from focus using multilayer feedforward neural networks
- 1 November 2001
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 10 (11), 1670-1675
- https://doi.org/10.1109/83.967395
Abstract
The conventional shape-from-focus (SFF) methods have inaccuracies because of piecewise constant approximation of the focused image surface (FIS). We propose a scheme for SFF based on representation of three-dimensional (3-D) FIS in terms of neural network weights. The neural networks are trained to learn the shape of the FIS that maximizes the focus measure.Keywords
This publication has 7 references indexed in Scilit:
- Three-dimensional shape recovery from the focused-image surfaceOptical Engineering, 2000
- Depth from defocusing using a neural networkPattern Recognition, 1999
- A variational approach to recovering depth from defocused imagesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Learning shape from shading by a multilayer networkIEEE Transactions on Neural Networks, 1996
- Accurate recovery of three-dimensional shape from image focusIEEE Transactions on Pattern Analysis and Machine Intelligence, 1995
- Shape from focusIEEE Transactions on Pattern Analysis and Machine Intelligence, 1994
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989