Tarımsal ürünlerin çarpma sesi kullanılarak sınıflandırılması

The quality is the main factor that directly affects the price for many agricul-tural produces. The quality depends on different properties of the produce. Mostimportant property is associated with health of consumers. Other propertiesmostly depend on the type of concerned vegetable. For instance, emptiness is im-portant for hazelnuts while openness is crucial for the pistachio nuts. Therefore,the agricultural produces should be separated according to their quality to main-tain the consumers health and increase the price of the produce in internationaltrades. Current approaches are mostly based on invasive chemical analysis ofsome selected food items or sorting food items according to their color. Althoughchemical analysis gives the most accurate results, it is impossible to analyze largequantities of food items.The impact sound signal processing can be used to classify these producesaccording to their quality. These methods are inexpensive, noninvasive and mostof all they can be applied in real-time to process large amount of food. Sev-eral signal processing methods for extracting impact sound features are proposedto classify the produces according to their quality. These methods are includ-ing time and frequency domain methods. Several time and frequency domainmethods including Weibull parameters, maximum points and variances in timewindows, DFT (Discrete Fourier Transform) coefficients around the maximumspectral points etc. are used to extract the features from the impact sound. Inthis study, we used hazelnut and wheat kernel impact sounds. The success rateover 90% is achieved for all types produces.iiiivKeywords: Impact sound, Pistachio nuts, Hazelnuts, Wheat kernels, Feature ex-traction, Classification, Food quality, Aflatoxin, Mel-Cepstrum, Principle Com-ponent Analysis (PCA), Support Vector Machines, Acoustics.