<![CDATA[Tangerine (Keprok) is one of well known horticultural commodities products in Indonesia. Tangerine fruit quality is determined manually, so far. The tangerine's skin texture can be used as an indication of the tangerine image quality determination. This study implements the Haar Wavelet method to determine the quality of tangerines in standard and automatic ways. The processes carried out in this study are preprocessing, Haar Wavelet feature extraction, and classification of tangerine fruit images using the k-NN algorithm. Statistical calculations of mean, standard deviation, and skewness are used to represent the Haar Wavelet features. The result of the Haar Wavelet extraction process is an image of tangerine that has been classified into Grade Super, Grade A, or Grade B. The Haar Wavelet decomposition process in this study was carried out at level 1 and level 2. The test results showed that level 1 Haar Wavelet decomposition produces a higher accuracy than Level 2. The highest accuracy obtained is 90%.]]>
Copyrights © 2020
