A method of recognizing and classifying 3-D shapes with continuous surfaces
by integrating shadow moire technique and neural network is presented. Unl
ike existing methods of 3-D shape recognition that use range images of poly
hedral objects or objects of different geometries such as cones, rods, sphe
res, etc., the proposed method classifies continuous surfaces that are geom
etrically similar. The objects selected to test the classification method a
re eggs of four different grades. The shadow moire technique, which has gre
ater sensitivity compared to structured lighting or laser scanning, is used
to obtain moire patterns on the surface of the eggs. From the moire patter
n images 14 parameters are extracted and used as input to a multilayer feed
forward neural network. The results of the classification using the neural
network show that the prediction accuracy attainable is 60% when classifica
tion is performed on all four grades. The accuracy increased to 95% when th
ree of the grades are classified. (C) 2001 Society of Photo-Optical Instrum
entation Engineers.