Classification of Hazelnut Kernels by Using Impact Acoustic Time-Frequency Patterns. (13th December 2007)

• Record Type: Journal Article
• Authors: Kalkan Kalkan, Habil Habil; Ince Ince, Nuri Firat Nuri Firat; Tewfik Tewfik, Ahmed H. Ahmed H.; Yardimci Yardimci, Yasemin Yasemin; Pearson Pearson, Tom Tom
• Other Names: Van hamme Van hamme Hugo Hugo Academic Editor.
• Is Part Of: EURASIP journal on advances in signal processing. Volume 2008(2008)
• Publication Date: 2007-12-13
• Subjects: Signal processing -- Periodicals; Traitement du signal; Signal processing; Periodicals; 621.3822
• DOI: 10.1155/2008/247643
• Languages: English
• ISSNs: 1687-6172

Abstract

Abstract : Hazelnuts with damaged or cracked shells are more prone to infection with aflatoxin producing molds ( Aspergillus flavus ). These molds can cause cancer. In this study, we introduce a new approach that separates damaged/cracked hazelnut kernels from good ones by using time-frequency features obtained from impact acoustic signals. The proposed technique requires no prior knowledge of the relevant time and frequency locations. In an offline step, the algorithm adaptively segments impact signals from a training data set in time using local cosine packet analysis and a Kullback-Leibler criterion to assess the discrimination power of different segmentations. In each resulting time segment, the signal is further decomposed into subbands using an undecimated wavelet transform. The most discriminative subbands are selected according to the Euclidean distance between the cumulative probability distributions of the corresponding subband coefficients. The most discriminative subbands are fed into a linear discriminant analysis classifier. In the online classification step, the algorithm simply computes the learned features from the observed signal and feeds them to the linear discriminant analysis (LDA) classifier. The algorithm achieved a throughput rate of 45 nuts/s and a classification accuracy of 96% with the 30 most discriminative features, a higher rate than those provided with prior methods.