Title :
Shape from focus using multilayer feedforward neural networks
Author :
Asif, Muhammad ; Choi, Tae-Sun
Author_Institution :
Signal & Image Process. Lab., Kwangju Inst. of Sci. & Technol., South Korea
fDate :
11/1/2001 12:00:00 AM
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 :
feedforward neural nets; image representation; SFF; focus measure; focused image surface; multilayer feedforward neural networks; neural network weights; piecewise constant approximation; representation; shape-from-focus methods; three-dimensional FIS; Cameras; Computational efficiency; Feedforward neural networks; Focusing; Geometry; Lagrangian functions; Multi-layer neural network; Neural networks; Polynomials; Shape measurement;
Journal_Title :
Image Processing, IEEE Transactions on
