Title :
Dishes extraction with neural network for food intake measuring system
Author :
Fujiwara, Koichi ; Takeda, Fumiaki ; Uchida, Hisaya ; Lalita, Sakoobunthu
Author_Institution :
Dept. of Inf. Syst. Eng., Kochi Univ. of Technol., Japan
Abstract :
We describe a food intake measuring system. This system measures the amount of food intake of each dish. which is put on the tray. Therefore, this system needs to extract dish image from the tray. In this paper, we propose a neural network algorithm (NN) for the dish extraction process. We expect the proposed method can extract the dish image accurately, even if the dish is over lapped partially with other dishes. We develop the mechanism of dish extraction by the NN and simulate it by using real data to show its effectiveness in practical use.
Keywords :
computer vision; feature extraction; health care; neural nets; patient monitoring; computer vision; dish image; feature extraction; food intake measuring system; learning conditions; neural network; Data mining; Food technology; Image converters; Image databases; Neural networks; Pixel; Shape;
Conference_Titel :
SICE 2002. Proceedings of the 41st SICE Annual Conference
Print_ISBN :
0-7803-7631-5
DOI :
10.1109/SICE.2002.1196556