<![CDATA[In order to improve the recognition rate and accuracy of stored grain pests classification, saving classification time, a new recognition method based on the Gabor wavelet and sparse representation is proposed in this paper. In this paper, nine typical pests in the stored grain are regarded as the research object, Gabor energy features and morphological features are extracted, principal component analysis is used to reduction dimension and sparse representation is used to achieve the classification of stored grain pests. Simulation results show that, Gabor energy feature is a better choose for grain pests classification, and the overall performance of Gabor features and sparse representation is better than the traditional classification methods. DOI : http://dx.doi.org/10.11591/telkomnika.v12i6.4860]]>
Copyrights © 2014
