Wood Texture Analysis by Combining the Connected Elements Histogram and Artificial Neural Networks

Authors:
Miguel Angel Patricio Guisado;Darío Maravall Gómez-Allende
Affiliations:
-;-
Venue:
IWANN '01 Proceedings of the 6th International Work-Conference on Artificial and Natural Neural Networks: Bio-inspired Applications of Connectionism-Part II
Year:
2001

Citing 4
Cited 2

A Theory for Multiresolution Signal Decomposition: The Wavelet Representation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Ten lectures on wavelets

Ten lectures on wavelets
Hierarchical classification of surface defects on dusty wood boards

Pattern Recognition Letters
Segmentation of Text and Graphics/Images Using the Gray-Level Histogram Fourier Transform

Proceedings of the Joint IAPR International Workshops on Advances in Pattern Recognition

Automatic visual inspection of wooden pallets

IEA/AIE'2003 Proceedings of the 16th international conference on Developments in applied artificial intelligence
A novel generalization of the gray-scale histogram and its application to the automated visual measurement and inspection of wooden Pallets

Image and Vision Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The automatic analysis of wood texture, based on a novel concept: the Frequency Histogram of Connected Elements (FHCE) is the main contribution of this work. The FHCE represents the frequency of occurrence of a random event, which not only describes the texture's gray-level distribution, but also the existing spatial dependence within the texture. The exploitation of the FHCE's shape, alongside its wavelet transform, allows the computation of excellent features for the discrimination between sound wood and defective wood; in particular, for the really hard pattern recognition problem of detecting cracks in used wood boards. A feedforward multilayer perceptron, trained with the backpropagation algorithm, is the specific ANN classifier applied for the detection and recognition of cracks in wood boards. A large digital image database, developed after an industrial project, has been used for testing purposes, attaining a success ratio far beyond those obtained with more conventional texture analysis and segmentation techniques.