- 學位論文
使用短距深度攝影機之用餐估算方法的研究
The Study of Food Intake Estimation Method Using Short-Range Depth Camera
若您是本文的作者，可授權文章由華藝線上圖書館中協助推廣。
摘要
在現代人們的生活中飲食評估是非常重要的，利用飲食評估系統來測量食物的重量，以評估該次用餐的攝取量是否充足。台灣目前為高齡化的社會，許多患有慢性疾病的老年人，經常透過飲食控制來抑制慢性疾病的復發次數，如何有效率並且準確地測量食物的重量，為本論文的研究主軸。 常見的量測方法是透過秤重的方式，將餐盤上的菜餚分別取出後，再單獨進行量測，而這樣的量測方式會耗費許多時間。因此，為了方便測量用餐攝取量，本論文以近距離使用深度攝影機，針對餐盤上的各道菜餚估算出飲食攝取量。在使用深度攝影機的過程中，發現兩個會影響本研究結果的問題，其一為餐盤材質對於深度攝影機的影響，其二為深度影像上有無深度值的像素點。本研究的流程步驟，首先分別記錄空餐盤、用餐前、用餐後的深度影像，接著分別估算出用餐前與用餐後的估算體積，最後使用比重函數轉換為重量。針對每一種食物，可以使用線性、二次或三次方程式來計算比重，求得近似的重量值。在實驗研究中，食物攝取量最低的平均誤差為5.6克，大約是7.5%。因此，本論文所提出的方法可以有效的估算出食物攝取量。
並列摘要
Dietary assessment is very important to people in modern life. Use the dietary assessment system to measure the weight of the food. To assess whether the intake of the meal is sufficient. Today, Taiwan is an aging society and many old people suffer from chronic diseases which always control the recurrence of chronic diseases by diet control. How to effective and accurate measurement of food weight is the main research objective of this paper. In general, food intake measurement is performed by using a scale. However, it is inconvenient since the food is measured one-by-one before and after eating. In order to increase the convenience of food intake measurement, a short-range depth camera is used to estimate the food intake directly on food tray. Two main problems are encountered to achieve the above goal. One is the influence of the material of the tray. The other is the pixel without depth values in the depth image. The volume is estimated firstly and then converted to the weight using the specific gravity function. For every food, its specific gravity function can be approximated using linear, quadratic, or cubic functions. In the experimental study, the lowest average error of food intake weight is 5.6 grams. It is about 7.5 percent. The results show that the proposed method can estimate the food intake weight effectively.
參考文獻
[1] M. H. Rahman et al., "Food Volume Estimation in a Mobile Phone Based Dietary Assessment System," in 2012 Eighth International Conference on Signal Image Technology and Internet Based Systems, 2012, pp. 988-995.
[7] F. Kong and J. Tan, "DietCam: Regular Shape Food Recognition with a Camera Phone," in 2011 International Conference on Body Sensor Networks, 2011, pp. 127-132.
[8] K. Khoshelham and S. O. Elberink, "Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications," Sensors, vol. 12, no. 2, 2012, pp. 1437-1454.
[9] J. Shang et al., "A mobile structured light system for food volume estimation," in 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2011, pp. 100-101.
[10] A. Telea, "An Image Inpainting Technique Based on the Fast Marching Method," J. Graphics, GPU, & Game Tools, vol. 9, no. 1, pp. 23-34, 2004.