Tarımsal uygulamalar için sinyal ve imge işleme algoritmaları

Medical studies indicate that acrylamide causes cancer in animals and certaindoses of acrylamide are toxic to the nervous system of both animals and humans.Acrylamide is produced in carbohydrate foods prepared at high temperaturessuch as fried potatoes. For this reason, it is crucial for human health to quanti-tatively measure the amount of acrylamide formed as a result of prolonged cook-ing at high temperatures. In this thesis, a correlation is demonstrated betweenmeasured acrylamide concentrations and NABY (Normalized Area of BrownishYellow regions) values estimated from surface color properties of fried potato im-ages using a modified form of the k-means algorithm. Same method is used toestimate acrylamide levels of roasted coffee beans. The proposed method seemsto be a promising approach for the estimation of acrylamide levels and can findapplications in industrial systems.The quality and price of hazelnuts are mainly determined by the ratio of shellweight to kernel weight. Due to a number of physiological and physical disorders,hazelnuts may grow without fully developed kernels. We previously proposed aprototype system which detects empty hazelnuts by dropping them onto a steelplate and processing the acoustic signal generated when kernels hit the plate. Inthat study, feature vectors describing time and frequency nature of the impactsound were extracted from the acoustic signal and classified using Support VectorMachines. In the second part of this thesis, a feature domain post-processingmethod based on vector median/mean filtering is shown to further increase theseclassification results.Keywords: Acrylamide, fried potatoes, coffee, k-means, image analysis, color, seg-mentation, median/mean filtering, hazelnuts, acoustics, classification, aflatoxin.iii