Design of an automatic wood types classification system by using fluorescence spectra
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
The complete follow-up of products from its originto its final use has been strongly developed in the lastyears. In wood industries, the implementation ofordinary identification systems presents implantationproblems, mainly because of the extremely variablenature of the material and the particular features of themanufacturing process. In order to allow follow-up ofproducts, new solutions have been considered by theuse of non-destructive control techniques. Weimplemented an identification system, where eachproduct is considered as unique with unique features.For this research we use a microwave sensor to obtainan intrinsic signal of a wood piece and then perform itsidentification. The objective is to determine a processfor identifying signals. The algorithm for patternidentification developed here is based on the k-nearestneighbor method. To increase performance of thisalgorithm, the signals are pre-treated. Signals alreadyrecorded in the database are positioned in the space ton-dimensions. When a product is to be identified it willalso see its signal positioned; the nearest signal havingas a reference the distance criterion will be located.The identifying algorithm developed shows apercentage of error of 1,5%.