<![CDATA[There are some problems regarding food that occur. One of them is nutrition and the quality of food that still needs attention. To find out the nutrients content in food we can use food classification using digital images. Classification requires an initial process called segmentation. In this case, the color space that used is Hue Saturation Value (HSV) and Otsu thresholding. Segmentation in this thesis uses 50 traditional cake images, begins with converting RGB images to HSV images. The Otsu thresholding is performed on each color component. Based on the results of these studies, the Value component of color gives the opposite result, the background is white and the foreground is black. Therefore, invert is applied to it. After thresholding on each color component, accuracy, specificity and sensitivity are obtained. Hue color component has an average accuracy rate of 42.64%, Saturation color component has an average accuracy rate of 94.34%, Value color component has an average accuracy rate of 70.68%. Tests for specificity and sensitivity show that Saturation color component has a higher value than other color components, with values 82.08% and 91.30%. Thus the Saturation color component is best used for segmentation using Otsu Thresholding.]]>
Copyrights © 2019
