We have applied color and texture segmentation in [2]. We applied our method to 3 different categories of food: single food, non-mixed food, and mixed food. In [ 2], we achieved better accuracy for non-mixed food, and our method had problems in detecting some mixed foods. As an example, illumination of food portions in a mixed food may change as they get mixed, making it harder to extract different food portions. Furthermore, the size of food portions in different mixed food are not similar, hence the method fails to segment food portions properly. To solve these problems, in this paper we have applied graph cut segmentation to improve our segmentation step. We also trained the system with more mixed food to get more accurate results.