An imaging technique for acrylamide identification in potato chips in wavelet domain

摘要

Acrylamide is a toxic compound which is commonly found in heated starchy food items like potato chips, French fries among others. Conventional methods based on chemical analysis for acrylamide identification are destructive, time consuming & expensive. This paper presents an efficient image processing based non-destructive method for identification of presence of acrylamide from potato chips based on statistical analysis and classification of the image features in wavelet domain. This proposed work involves segmentation of the area of the potato chip from the background of the image followed by feature extraction in wavelet domain. These wavelet features are strategically analyzed for acrylamide identification using k-nn (k-nearest neighbors) based classification method. The experimental results obtained using the proposed method indicates that wavelet features have discriminatory properties for identification of presence of acrylamide in food items. The experimental results of 97% accuracy and 100% sensitivity could be explored for commercial use. (C) 2015 Elsevier Ltd. All rights reserved.